The Minneapolis Plan to end Too Big to Fail

Like our favorite son of that same city, it is not lacking for audacity: a plan to end Too Big to Fail (TBTF) put forward by the Federal Reserve Bank of Minneapolis. The engineering of the Minneapolis Plan is based on a simple, straightforward underlying assumption: more capital means lower risk of default. The Plan calls for the imposition of capital standards considerably stricter than what US regulators, under the aegis of Basle 3, have so far required for even the biggest and worldliest behemoths, the G-SIBs (Global Systemically Important Bank). There’s a kicker in the package, too, a capital sin tax — a corner solution for systemically important banks so onerous that it is unsustainable, forcing those entities to shrink to systemically unimportant proportions. And if those proposals were not bold enough, try this on for size: the Plan declares a major plank of the US bank regulatory agenda — the issuance of long-term debt subject to conversion to equity (summarized by Total Loss Absorbing Capacity, or TLAC) — to be temperamentally unfit for deployment.

There are elements of the Minneapolis Plan we admire. It offers the prospect of a simpler regulatory regime for community banks. We, too, think TLAC is ill-conceived, and prefer the surety of high quality capital versus the fear-inducing, squidgy, and possibly destabilizing convertible option. But as for the wisdom of forcing the G-SIBs to cut down to size so they can pass through the Orderly Liquidation Authority (OLA), more study is needed. It is noteworthy that the Minneapolis document does not discuss banking industry organization and scale in depth. Apparently the Big of TBTF speaks for itself; Bigness brings unacceptable macroeconomic risks. It is hard to find fault with the concept of a resolution regime, though the prospect of its actual use is not without risk, potentially adding to the social cost burden that figures in the Minneapolis analysis. We look to expand on these matters in future posts.

We are in favor of capital requirements higher than what was in place prior to the Noughts Panic. How much higher is difficult to say. The quality of the data and analysis on historical losses and capital write-offs is, dare we say it, poor. Concepts of capital management seem to be badly understood by many, ourselves included. We will take up this point in a later post in this series. We do not believe that it is practical to set capital levels high enough to induce near-zero likelihood of government intervention without major restrictions in credit and risk transfer flows. It is hard to imagine what such banks would look like. All that can be said is that the marginal — the smallest, the riskiest, the new entrants, the ones closest to the perimeters — are those that would be denied credit. The dynamism of the US economy, already in peril, would recede further.

We can see an end to TBTF, but we refer to the demise of the phrase TBTF, not the policy itself. So much emotion is hung on that phrase, and it needs to be buried: it takes up too much airtime, it’s sloganeering at its very worst. Other countries such as Canada carry the policy, quietly, without the baggage of the notorious tag. We believe the unwarranted focus on the policy distorts proper considerations of safety and soundness. Do construct policy to minimize the application of TBTF, but there are other far more pressing problems in banking policy that should take precedence over its elimination. Why eliminate options when policy may be able to get close to the goal, and there is little to gain economically from its eradication?

The frequent use and abuse of the phrase TBTF to us indicates the presence of gaps in the conceptual scaffolding of banking and finance: proper consideration of network effects, industry organization and risk. We do not believe banks can be effectively blocked off from excessive risk elsewhere in the financial system. Contagion and connectedness are recognized in the literature, but why not their ubiquity and persistence? Is it that capital, collateral, liquidity requirements, the discouragement of repo, and the promotion of central counterparty clearing are seen as effective cleansers of contagion? That these structural reforms will limit the movement of asset prices? That risk will be tamed to such an extent that “offered only” situations are history?

A classic exposition of market connectedness can be found in the swoon of US markets in August/September 1998. We recap the story:

Emerging Asian markets, beginning with Thailand, had been battered since the summer of 1997, and the sour sentiment drifted over to other emerging markets in classic Kindlebergian fashion. Latin American markets flopped in the second quarter of 1998, the Brazilian Bovespa index falling by 26 percent in two months beginning April 15, the Mexican IPC index dropping 22 percent in the same period. A flight-to-quality response resulted in a significant rally in US Treasuries (UST), with 10-year UST yields falling 150–200 basis points since mid-1996. A follow-on prepayment surge in mortgage-backed securities left a dent in the returns of a number of hedge funds in the first half of 1998, including, most famously, Long-Term Capital Management (LTCM). The August 17, 1998 default of Russia threw already wobbly US markets into turmoil. The stress that was felt is only somewhat apparent from market soundings: the S&P 500 fell 20 percent in six weeks beginning in mid-July; 5-year swap spreads transited from the +mid-40s in early August to +98 basis points in mid-October, but LIBOR betrayed very little strain. A number of markets were ultimately affected, including asset-backed securities and older vintage USTs.

The securities business was walloped by the turn of events, and risk tolerance shifted swiftly and profoundly. A BIS-sponsored study by the Working Group on Financial Market Events in the Autumn of 1998 concluded that not only were Value at Risk (VaR) models found to be deficient under extreme conditions, but contagion was amplified by the unwinding of positions that were outside VaR limits. Some banks bore considerable trading exposures to emerging market debt, others were sucked in through fixed-income positions and exposure to hedge funds. The announcement of trading losses in a deteriorating market led to fears for the solvency of financial institutions. On the day the merger between Travelers and Citicorp was consummated, October 8, 1998, the new Citigroup’s Salomon Smith Barney unit disclosed USD 700 million in trading losses. Hello, world! Lehman Brothers and Bankers Trust were compelled to publicly defend their solvency[1]. Richard Fuld’s successful fight for the survival of Lehman surely affected his approach to the firm’s management in 2008. Lehman’s stock had fallen 70 percent, and it was reported at the time, foreshadowing events of a decade later, that the firm gave information about short positions in its shares to the SEC[2].

Hedge funds, especially those exposed to mortgage-backed securities, were subject to margin calls[3]. The temporary recapitalization and wind-down of LTCM in September was a signal event. Reminiscent of clearinghouse arrangements of a bygone era, the actions were undertaken by a 14-member bank/dealer co-operative, at the direction of the Federal Reserve Bank of New York, to ensure an orderly unwind of the LTCM portfolio. Bank and broker/dealer exposure to hedge funds unsurprisingly became an item of concern to investors. An astonishing revelation came from Bank of America regarding an arrangement with the hedge fund D.E. Shaw[4].  Bank of America wrote off USD 372 million of a USD 1.3 billion loan to Shaw and agreed to buy Shaw’s entire USD 20 billion portfolio of fixed income securities and associated hedges, which was to be sold off with proceeds applied against the loan.

Trouble encroached on new issue securities and funding markets. Warehouse funding lines ― short-term funding provided by broker/dealers and banks for non-conforming real estate loans ― were pulled, causing the bankruptcies of two mortgage lenders, Southern Pacific Funding Corporation and Criimi Mae, in October. The term asset-backed securities market was essentially shuttered for a number of weeks that fall, the first time that had happened since the market’s inception in the mid-1980s, though asset-backed commercial paper was accommodating[5]. Investment grade and high yield corporate bond issuance too were delayed.

The Federal Reserve had been on hold since a 25 basis point hike in Fed funds to 5.50 percent in March 1997. Fed funds was eased by a total of 75 basis points in September, October, and November 1998.  Chairman Greenspan, on an October 15 FOMC conference call[6], said (p. 29): “At this stage, after 50 years of looking at the economy on almost a daily basis, I must say that I have never seen anything like the current situation. Certainly, based on all the historic annals I have read, and I have done a good deal of reading in economic history, it would be an extremely rare event for this type of financial environment to emerge and eventually to recede without having any impact on the economy. Indeed, I do not remember any occasion when that occurred in the past.” Extremely rare, perhaps, but we prefer to categorize the contingency inherent in the progression of such events. The turmoil did recede, a break appearing after the October 15 intermeeting ease. That move seemed to summon the “confidence fairy,” a nice turn of phrase employed by Joseph Stiglitz in his 2010 book Freefall. Primary markets resumed flows directly afterward, and many secondary markets regained order. Some tentativeness remained, some industry landscape shifted in the wake of the near-miss. Assets of US broker/dealers shrunk seven percent in Q4 1998 and another three percent in Q1 1999, but growth resumed in Q2 1999[7].  Commercial bank assets, however, grew throughout the period. Some banking personnel moves were made, and a weakened Bankers Trust, which had lost USD 488 million in Q3 1998, was acquired by Deutsche Bank, the largest foreign acquisition of a US bank to date. A number of economists forecasted recession in the wake of the crisis, but the US expansion continued until March 2001. The Federal Reserve resumed rate hikes in June 1999.

Would higher capital requirements have forestalled the 1998 event? It is impossible to say. Sure, on paper, in a comparative statics analysis, for a given book of business a higher level of capital will lower risk of default. Perhaps Bankers Trust would have lived to see another independent day. But there is so much more to consider, both inside and outside any given banking institution. Banking is completely dynamic; numerous decisions are made daily. How should costly capital be deployed? It is entirely plausible that, back in the mid-/late-1990s, meaningfully higher capital requirements might still have led managers to favor loans to hedge funds, collateralized on the funds’ security holdings; the rationale being, besides the collateral, there was a good degree of comfort with the loans given familiarity with the fund business and personnel. Perhaps these loans were granted in order to get additional business from the firms. Emerging markets positions, too, might have been favored in a high capital regime given their high expected returns. Those were heady days of globalization: the Asian Tiger story was alluring, Russia was still fresh out of the blocks as a nation in transition.

It needs to be kept in mind that, just as for nonfinancial firms, bank equity is considered currency of the firm. It figures into remuneration. It can be used for acquisitions. A growing share price signifies the firm’s health. Stasis will not do; managers will drive forward the value of that currency however they can. If higher capital requirements are in place, then there is a nonzero likelihood that that the firm’s overall risk will actually increase, in order to boost returns and the value of the firm. But the likelihood that overall risk will increase depends on many factors, endogenous and exogenous, including the degree of regulatory permissiveness and demand for services. There are other factors determining risk tolerance as well, such as collateral, liquidity, competition and expectations. And never forget overall system liquidity, as determined by the Federal Reserve.

Still other events can affect asset valuations and therefore risk tolerance, as we have seen. Currency swings in Asian emerging markets in 1997 ended up in US Treasury, mortgage- and asset-backed markets. The Federal Reserve should be commended for taking a very bold move in September 1998 to help preserve orderly markets. Wags waxed snarky about the “Greenspan put” but that Federal Reserve regime understood the economic value of nongapping markets, the preservation of wealth, and the critical role of the confidence fairy. Think of how interest rates have traveled in the last 20 years, how that has affected asset prices, and the rate risks that lie in the weeds.

High capital requirements are not enough to eliminate the risk of unhinged markets. One of the things we liked the most in the Minneapolis Plan paper (p. 43) was the report of an audience member at a symposium suggesting that a goal of reform should be “…a system that prevents the premature and inefficient liquidation of valuable assets”. The 1998 market event clearly demonstrates how contagion can be set off by pre-existing protective mechanisms, such as VaR and margin calls. The recovery of prices may come too late for some institutions. The social benefit of the Greenspan put should be acknowledged and studied further.

We acknowledge the risk of social cost inherent in our global financial system, which leads us to ask whether a higher price should be paid for insurance, and how much society benefits on a long-term basis from the risks that have been taken.



[1] Daniel Dunaief and Ellen Taylor, “Bankers Trust Denies Trouble as Shares Fall (Update 1),” Bloomberg, October 16, 1998.  Lisa Kassenaar, “Lehman Hedge Fund, Emerging Market Risk Is $236 Mln (Update 2).” Bloomberg, October 5, 1998.

[2] Dylan Gallagher, “Lehman Gives SEC, NYSE Information on Short-Sellers, WSJ Says,” Bloomberg, October 5, 1998.

[3] The quick resurrection of markets led to hedge fund lawsuits against claims of hasty and unnecessary forced securities sales.  Andrew Galvin, “Ellington Hedge Funds Sue UBS Warburg Over 1998 Asset Sales,” Bloomberg, June 18, 2000, and Katherine Burton, “Salomon Faces Possible Suit by Hedge Fund Over 1998 Margin Call,” Bloomberg, October 25, 2000.

[4] Daniel Dunaief, “BankAmerica Profit Falls 50% on Trading, Loan Losses (Update7),” Bloomberg, October 14, 1998.

[5] The reverse happened in 2007.  The term ABS market was accessible for many issuers and asset classes after July 2007 (with some exceptions, most notably subordinated securities and home equity, but even there, shutdown was partial until 2008), while a considerable amount of ABCP had difficulty rolling.

[6] Board of Governors of the Federal Reserve System, FOMC: Transcripts and Other Historical Materials, 1998., accessed January 11, 2017.

[7] Board of Governors of the Federal Reserve System, Z.1 Financial Accounts of the United States.


Admati, Anat R., Peter M. DeMarzo, Martin F. Hellwig, and Paul Pfleiderer. 2013. Fallacies, irrelevant facts, and myths in the discussion of capital regulation: why bank equity is not socially expensive. October 23.

Federal Reserve Bank of Minneapolis, 2016. The Minneapolis Plan to end Too Big to Fail. November 16.