Runnable, Ruinable Repo and the Great Recession: a Panic-proofing Approach to Financial Regulation

JurisdictionUnited States,Federal
Publication year2018
CitationVol. 5 No. 2

Runnable, Ruinable Repo and the Great Recession: A Panic-Proofing Approach to Financial Regulation

Aaron Metviner

RUNNABLE, RUINABLE REPO AND THE GREAT RECESSION: A PANIC-PROOFING APPROACH TO FINANCIAL REGULATION


Aaron Metviner*


We want a system that is not prone to panics, or, better yet, a system that does not have panics, a system where losing confidence does not happen . . . In the aftermath of the [2008 financial crisis], how can such a system be designed?
—Gary Gorton, Slapped in the Face by the Invisible Hand (2010)


Introduction

"The [2008 financial] crisis," then Federal Reserve Chairman Ben Bernanke reflected at a 2012 conference on financial regulation, "is best understood as a classic financial panic."1 A "panic" occurs when short-term debt claimants decide en masse to no longer fund the banking system.2 Yale economist Gary Gorton has shown that "the most important part of the panic occurred in the repo market."3 "Run on repos," he holds, "[was] the core problem in the financial crisis."4

"Repo" is shorthand for "sale and repurchase agreement." Institutional financiers—like money market mutual funds, pension funds, and corporate treasuries—enter into these short-term (typically overnight) "repo agreements" with Wall Street firms. Repo provides them with a sort of demand deposit

[Page 86]

account for their excess cash.5 It also gives Wall Street firms access to an enormous amount of cash. The "run on repo" was a run by major financial institutions (the repo lenders) on other major financial institutions (the repo borrowers).

Demand deposit accounts and overnight repo agreements serve a similar function for retail depositors and repo "depositors" respectively. Large financial institutions, not unlike regular people, want to earn interest on their excess cash while not losing the ability to access their cash at any time. That is why they enter into these overnight repo agreements with Wall Street banks. Each day, they can either "withdraw" their cash or "roll over" their repo. If they choose the former, the repo borrower—under the terms of the repo agreement—must obviously honor that demand.

The key practical difference between demand deposits and overnight repo "deposits" is that the federal government provides deposit insurance for the former but not the latter.6 To lessen the repo lenders' counterparty default risk (i.e., that the Wall Street bank won't have the cash on hand to honor the repo lender's "withdrawal"), repo deposits are collateralized. Repo lenders are fully secured creditors of their counterparty banks, contractually armed with immediate recourse to the specified collateral under each repo agreement if their counterparty bank defaults.

One of the key lessons to be drawn from the 2008 financial crisis is that this security interest was not enough of a security interest to offset the repo lenders' counterparty default risk. In other words, having immediate recourse to collateral—even high-quality collateral—was inadequate to temper the concerns of skittish repo lenders and contain the contagious run on repo.7 Vanderbilt law professor and former Treasury bureaucrat Morgan Ricks has made this point loud and clear, most recently in his 2016 book The Money Problem.

Gary Gorton and Morgan Ricks are two of academia's main proponents of the "panic-proofing" approach to financial regulation. That approach, on which

[Page 87]

this paper ideologically stands, is based on the following logical progression and understanding of US banking history:

1 - US banking history was marked by frequent financial panics8 (and in turn macroeconomic tragedies) until the Depression-era advent of deposit insurance in 1934. Deposit insurance is to thank for the 75-year "Quiet Period"—of no systemic panics from the Depression to the Recession—in US banking history. 9

2 - A panic by institutional financiers, mainly overnight repo creditors on systemic financial institutions,10 commenced in August 2007.11 This panic reached its acute phase in September 2008, when twelve out of America's 13 leading financial institutions faced dire threats to their survival.12 If not for this panic, the Great Recession would have been far less severe and might not have happened.13

3 - Systemically important financial institutions (the "SIFIs," mostly Wall Street banks14 ) continue to rely heavily on runnable short-term debt to finance their long-term activities. Thus, "panics represent far and away the biggest threat that the financial system poses to the broader economy."15 The focus of financial regulation should therefore be on "panic-proofing."16

[Page 88]

Ricks and Gorton agree that the focus of financial regulation should be on panic-proofing. They have different ideas, however, on how policymakers should go about panic-proofing the financial system. Gorton would reform the repo market by ensuring the high quality of repo collateral. Ricks would federally insure all the short-term debt obligations of systemically important commercial banks. And both would limit the exposure levels of the SIFIs to runnable short-term debt.

This paper argues for a sort of Gorton/Ricks hybrid approach to panic-proofing, whereby the federal government (1) insures the very-short-term repo obligations of the SIFIs (while also ensuring the high-quality of the collateral under those repo agreements) and (2) limits the exposure levels of the SIFIs to all other forms of uninsured runnable debt.

Section I will discuss the historical panic-proofing success of deposit insurance in the US. It will explain how deposit insurance became an anachronistic solution to forestalling panics, given the rise of the money markets and the institutional "depositor." It will then cover the repo market, the run on repo of 2007 and 2008, and why repo lenders ran en masse from the SIFIs in the prelude to the Recession.

Section II will examine the two main pieces of financial legislation to emerge from the Great Recession, the domestic law Dodd-Frank and the international accord Basel III. It will specifically ask what these laws do to contain panics by repo lenders, as well as discuss their general "risk-constraint" (not panic-proofing) approach to financial regulation.

Section III will expand upon the panic-proofing models for financial regulation as proposed by Ricks and Gorton. It will critique these models before offering this paper's original panic-proofing approach.

I. RUNNABLE, RUINABLE REPO AND THE GREAT RECESSION

A. Banking's Inherent Susceptibility to Panics

The classic business model of banking involves a heavy reliance on short-term debt to finance long-term assets.17 This almost has to be the case, since

[Page 89]

"part of making the "deposit" nearly riskless is for it to be short maturity."18 But this makes banks "inherently susceptible to a liquidity crisis or "run" in which short-term claimants simultaneously seek to redeem."19

Prior to the Depression-era establishment of deposit insurance in 1934, American banking was marked by regular panics and macroeconomic storms. Concerned their banks could fail, retail depositors rushed to withdraw their funds. Failure to withdraw in time could mean becoming an unsecured creditor of a failed bank in a bankruptcy proceeding—and perhaps having to wait years before receiving any money back.

B. Establishment of the FDIC and the "Quiet Period" in US Banking

The tragedy of the Great Depression, triggered by a widespread panic by retail depositors, prompted the US Congress to establish federal deposit insurance in 1934. A new federal agency, the Federal Deposit Insurance Corporation (FDIC), would now insure all bank deposits at insured member banks up to a certain dollar amount (today $250 thousand).

Retail depositors could now rest assured that, if their bank were to fail, the FDIC would promptly send them a check covering the full amount of their deposit balance. They now had no reason to withdraw their funds from their banks in a frenzy anymore—their deposits were backed by the full faith and credit of the US Treasury.

The advent of deposit insurance set the stage for the 75-year "Quiet Period" in American banking—a period marked by no systemic panics.20 Deposit insurance "stopped the cycle of runs on demand deposits and allowed them to be used safely as money,"21 creating an "unusually stable time for the US economy."22

C. Rise of the Institutional "Depositor" and Unraveling of the Quiet Period

The Quiet Period began to unravel in the 1980s. Wall Street firms, historically reliant on retail deposits for their cash needs, began to rely

[Page 90]

increasingly on institutional money—that of money market funds, pension funds, and other large financial institutions. There had been an "explosion of the modern day money markets,"23 and these financiers needed a safe place to store their excess cash in the short-term.24 Repo provided that place. Retail deposits, in turn, diminished in importance as a financing vehicle for Wall Street firms.

By 2007, a run by these institutional repo creditors would be capable of leading to large-scale liquidity problems on Wall Street. On Wall Street's collective pre-Recession balance sheet, liabilities for repo debt far exceeded the banks' readily available liquidity.25 And repo deposits were not FDIC-insured (they well exceeded $250 thousand in value). Gorton states the obvious point: "Deposit insurance works well for retail investors but leaves a challenge for institutions with large cash holdings."26 FDIC insurance no longer served a panic-proofing purpose.27

Commencing in August 2007—and culminating in September 2008 following the collapse of Lehman Brothers—an en masse run by repo creditors on their Wall Street counterparties brought the Quiet Period to a harsh close.

D. Repo

The "repo agreement"28 would overtime become an increasingly important vehicle of short-term debt financing for Wall Street firms. By the time of the Crisis, repo had become "the most important form of...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT