Shadow Banking: The system itself is a systemic risk
Many bankers and politicians will continue to recite the same romantic story that the Dodd-Frank Act of 2010, passed years following the financial crisis of 2007-2008, has made the banks “more safe". That would be terrific news if risky bankers caused the global financial crisis, but they were not. The 2007-2008 financial meltdown highlighted the vast and complex system of non-bank financial intermediaries that provide sources of funding for credit creation, much like a traditional bank, without the regulations and insurance. The system has become referred to as the “Shadow Banking System”, originally by Paul McCulley in 2007 to express the unregulated nature of the nonbank financial intermediaries that embody the system. In many ways, the ideas expressed here are not outside of the realm of common knowledge within the finance world; however, the economic implications of these finance operations are rarely strung together in such a way that exposes major concerns of economic stability. Most of the available work has been completed by Zoltan Pozsar, Tobias Adrian, and Adam Ashcraft; although there is no sign that their work has influenced policy makers.
A brief overview of the development of our banking system in the traditional sense, in which banks were primarily originating loans with the intent to hold those very loans on their own balance sheet, proves to be helpful in understanding the risks that are associated with the originate-to-distribute system of banking, in which loans are removed from the balance sheet of banks and enhanced. The major components of the shadow banking system continue to be securitization and collateral intermediation.
Traditional Banking Model and the Development of Shadow Banking
Our financial system consists of commercial banks such as depository institutions that have access to the Federal Reserve Bank (The Fed) and the Federal Depository Insurance Corporation (FDIC). These regulatory institutions were established to prevent the cascading effect that would be set in motion at any loss in confidence of a bank’s ability to pay out customer deposits on demand; essentially, to prevent a “run” on the bank. This is an inherent feature of our fractional-reserve banking system, that which you can read my previous article for further information.
In the event that there is a negative liquidity shock in the banking system, The Fed, through the discount window, acts as the lender of last resort and lends freely against “good assets” (as collateral) at a high rate of interest. When the banking system is experiencing a negative liquidity shock it is associated with a distinct difficulty to convert illiquid assets (i.e. structured money market instruments such as asset-backed securities, commercial paper, etc.) into more liquid forms, such as cash. In normal times this type of conversion is performed regularly without any hesitation. As lender of last resort, The Fed has guaranteed depository institutions, through the discount window, that they needn’t be concerned that a liquidity crisis may manifest itself and thereby mitigate the self-fulfilling prophecy of a bank run. Indeed, the insolvency of one bank or suspicion of insolvency would be sufficient to compel customers to withdraw their deposits which is further mitigated by the FDIC. Additionally, banks are also encouraged to first borrow and lend amongst themselves, which is ultimately cheaper. The discount window is only available to commercial banks chartered by the Federal Reserve Bank and not available to insurance companies, investment banks, broker-dealers, or other institutional cash pools. The fact of the matter is that commercial banks and their customers are protected on this front, however, there is a growing number of non-bank financial intermediaries that are performing the same types of functions without these protections. Moreover, these intermediaries are undoubtedly affecting the real economy through their lending activities but they do not use cash or anything that you or I would find valuable. Imagine a totem-pole of “money” with cash at the top of and various “near monies” below it such as commercial paper, asset-backed commercial paper, etc. It surely begs the question if the Federal Reserve is effective to the extent that it can influence the real economy or avert the downside risks of a recession. (It’s a curious industry, working in finance; imagine having a job in which you go to the office every day and use other people’s money to create fake monopoly money in order to make yourself real money, that’s finance.)
The story of the shadow banking system is best understood initially through the lens of loan securitization. Securitization is the process through which loans are removed from the balance sheet of lenders and transformed into “money-like” debt securities purchased by investors (i.e. mortgage-backed securities, as one example). These debt securities take many forms, depending on which shadow bank we refer to. That among these are asset-backed commercial paper (ABCP), asset-backed securities (ABS), collateralized debt obligations (CDOs) and repurchase agreements (repos). Perry Mehrling, a professor at Barnard College, defined shadow banking most succinctly as money market funding of capital market lending. A different way to understand that is a form of credit intermediation; funding long-term risky assets using short-term “risk-free” instruments.
Traditionally, the banking system operated as an originate-to-hold system of lending. Under which, a bank would lend (originate) to a borrower with the intention of holding the loan on their balance sheet until maturity. Throughout the life of the loan, the lending institution intended to service the loan by collecting payments from borrowers. However, banks were restrained in their ability to fund these loans to borrowers to the extent that they had sufficient deposits. Indeed, this particular system of banking had bounded profit margins. The transformation of short-term deposit liabilities into long-term loans is known as maturity transformation and it is inherently risky, especially when the value of the assets created from short-term collateral begins to fall unexpectedly.
Deposits are not the only account that supports bank lending, however, which is what contributed to the institutional transition toward “originate to distribute” lending practices. Banks issue debt and equity to further capitalize their intermediation activities and to protect their customer’s deposits. For example, a bank has $100 of loans supported by $10 of equity, levered 10 times, if 10% of their loans default that would put their depositor's money at risk. Perhaps, more importantly, it would potentially put the bank out of business. The Federal Reserve imposes capital requirements upon banks based upon the amount of risk that they take; this makes it more expensive to lend to borderline creditworthy borrowers and far less likely to lend to entrepreneurs. Reconcile these requirements with the regulations that forced banks to take increasingly more risk on subprime loans and do the math.
As it turns out, there is one thing that banks hate more than risk-- it’s called costly regulations eating away at their profit margins, especially when they are inherently risk-averse. In the 1980s, loan defaults in Latin America paved the way for minimum capital requirements for banks based upon risk. Banks were compelled to remain competitive while meeting economic demand for financing. Banks began to sell its assets, at first focusing on corporate loans, by creating an investment vehicle to pool these more risky loans that were subjected to higher capital requirements. I like to picture a tiny office with a balance sheet in the middle of an otherwise empty room to visualize a “special purpose vehicle.” (SPV) These were off-balance sheet “storage units” that were bankruptcy-remote. Not to imply that these vehicles were “bankruptcy-proof”, but rather to distinguish between the bank assets that finance the SPV from the credit risks within the SPV. This is the first step in a long chain of loan securitization that remains true today.
The risky loans were pooled, each with its own cash flow from monthly payments by the borrower and subsequently divided into “tranches” that were sold to investors across the globe. This should make sense; if I have multiple cash flows from hundreds of borrowers it would be nice, but it would be nicer to sell these future cash flows for their present value discounted back to today if given the opportunity. This accomplished two important tasks; banks could mitigate costly capital requirements and investors could provide the funding for loans that were high demand in accordance with their risk preferences, which were previously being suffocated by risk-based capital requirements.
Another example; cell phone companies recently changed their business model to retail phones for their full retail price and the customer typically pays the phone in 24 monthly installments. These companies are unwilling to wait for each monthly payment so they securitize these future cash flows such that they receive the full retail price up front (at a discount) and subsequently deliver customer’s monthly payments to the finance company they sell their receivables to.
The Mechanics of Securitization
Maturity transformation, the funding long-term assets (mortgage loans) with short-term liabilities (deposits), has been a core operation of banks and banks have a considerable amount of expertise in properly evaluating the creditworthiness of borrowers. Maturity transformation remains in the present shadow system however through a more complex process that utilizes the securitization process (loans that are transformed into trade-able assets) and wholesale funding (funding that is conducted in capital markets through investments such as asset-backed commercial paper). In the shadow system money market mutual fund provide the function that depositors provided in the traditional system.
Risky, long-term loans (such as subprime mortgages) are transformed into seemingly risk-free, short-term, money-like instruments with stable net asset-values that are issued by MMFs with daily liquidity. For example, a finance company originates loans and transfers them to a warehouse such as a bankruptcy-remote vehicle (like a bank-sponsored SPV) which warehouses pools of these loans until they are transferred to a broker-dealer to be combined with other loans and structured into an asset-backed security (ABS). The initial loan originator would typically rely on commercial paper or medium-term notes. The warehouse might rely on asset-backed commercial paper, with the store of loans serving as the collateral for that funding. The broker-dealer structures and then warehouses ABS through its trading book and may fund this process through repurchase agreements (repos) or repo conduits. A money market fund could purchase the commercial paper or other short-term debt issued by either entity (either the loan originator or ABS warehouse). This chain can continue through further warehousing and blending of ABS using a bankruptcy-remote SPV to create a CDO. Securitization allows loan originators to sell pools of debt (such as mortgages, credit card receivables, or auto loans) to other institutions, thus transferring credit risk.
The Implications of Securitization
I find it rather important to emphasize that there is no such mechanism that could dice up risk and sprinkle it inter-temporally as well as among several certain agents. There is either risk or no risk; it is not in the nature of risk that it be manipulated in such a way. The transfer of risk, at least as it used here, is to mean that the hierarchy of credit quality, in the form of tranches, allows for investors to fund projects that which match their risk preference. Although all I will say about measuring risk is that it is a curious feature in statistics that the future is much more predictable than the present. With that said, risk-averse investors should read the prospectus and skim the report provided by Moody’s.
In finance, we refer to deposit insurance provided by the FDIC as a “credit put” because it enables depositors to sell their deposits at par value. We also refer to the Fed’s discount window as a liquidity put because it enables member banks to sell (pledged as collateral) assets in exchange for short-term funding. These features make traditional banking superior to the shadow banking system; the credit put validates a bank’s liabilities while the liquidity put sets a price floor on their assets. When you begin to conceptualize banking, it becomes more apparent that is truly some forms of alchemy. Everyone would like to believe that they are liquid, although they also know that this cannot be true for all market participants at the same point in time. This paradox was emphasized by the famous economist John Maynard Keynes regarding system-wide liquidity, (or infamous, depending on perspective), “For the fact that each individual investor flatters himself that his commitment is ‘liquid’ (though this cannot be true for all investors collectively) calms his nerves and makes him much more willing to run a risk.” Liquidity is simply an illusion without the put options provided by The Fed. This is also true for deleveraging; everyone cannot deleverage at the same time. In the case of shadow banks, remember that they do not have immediate access to these liquidity puts, rather they rely upon wholesale funding. When asset prices stopped rising prior to the financial crisis and collateral prices no longer supported the large number of illiquid assets created by the system (which were ultimately toxic), a run on the shadow banks was set in motion. Unfortunately, a run on the shadow banks produces a different type of contagion than a traditional run such as the Panic of 1907. Instead, a forced sale of assets begins to push the price of the asset lower and lower while simultaneously rapidly raises the cost of funding all shadow banks-regardless of their individual financial health-which further increases the demand for liquidity. Such was the case of Lehman Brothers in 2008, potential counter-parties were already aware of their exposures to these toxic securities and therefore unwilling to lend. Lehman found itself unable to fund its illiquid obligations and was unable to exit their positions at prices that would satisfy its equity requirements. When the equity account goes negative, it is game-over; the firm filed for bankruptcy as their insolvency became imminent.
This is the type of credit intermediation that collapsed the global financial markets through creating mortgage-backed securities, most of which were infected by subprime mortgages. Had the government not forced banks to lend to borrowers who had no money then the past decade may not have been stained by high-employment and falling wages. But that is beyond the scope of this post. Debt securitization is by no measure an inherently bad thing; in fact, it provides a net benefit when the market is able to measure and respond to risk. For some reason that is beyond me, voters actually believe that government regulators and politicians know what they are doing when they promulgate their rules. The topic above is complex and I have simplified it as best as I could the past month, but you would be wrong if you assume that politicians understand this material. This is only the beginning of the implications of shadow banking. I’ve so far discussed the role of securitization in shadow banking, the following post that I will finish shortly and provide a link to below discusses the role that collateral intermediation plays and how this may impact the real economy.
Pozsar, Zoltan Shadow Banking: The Money View Office of Financial Research Working Paper, No. 14-04, July 2, 2014.
Pozsar, Zoltan; Tobias Adrian Adam Ashcraft and Hayley Boesky Shadow Banking Federal Reserve Bank of New York Staff Reports, No. 458.