Banking Without the Too-Big-to-Fail Doctrine

Mr. Salsman is a banker in New York City and an adjunct fellow of the American Institute for Economic Research in Great Barrington, Massachusetts. This article is adapted from a speech delivered at a conference sponsored by The Federal Reserve Bank of Dallas, May 12-13, 1992.

Since the failure of Continental Illinois in 1984, the U.S. government has pursued a deliberate policy of bailing out large commercial banks deemed “too-big-to-fail.”[1]

The “too-big-to-fail” doctrine has arisen not simply because of the growing number of bank failures in the past decade, though indeed failures have increased. In fact, the doctrine’s historical origins go back much further than a decade. More than 40 years ago, a 1950 amendment to the Federal Deposit Insurance Act of 1934 introduced the “essentiality doctrine.” As codified, that doctrine states that in its sole discretion the government can rescue any failed bank when “continued operation of such bank is essential to provide adequate banking service in the community.”[2] None of the key terms in that provision—such as “essential,” “adequate,” or “community”—has ever been defined, permitting arbitrary discretion to rule. Coupled with the diminishing financial condition of banks in subsequent decades, the “essentiality doctrine” has given government wide latitude to bail out failed or failing banks for whatever reasons it deems necessary.

Of course, deposit insurance legislation itself arose out of the bank failures of the early 1930s. These failures in turn were largely the result of Federal Reserve monetary mismanagement.[3] In short, today’s “too-big-to-fail” doctrine can trace its roots to the very establishment of central banking in this country in 1913. Before we examine the merits of the manner in which government has decided to handle bank failures, it is helpful to understand why banks are failing today in such large numbers to begin with.

The main theme of my own research on U.S. banking history has been that central banking is detrimental both to sound money and safe banking. In particular I have found that the U.S. commercial banking industry has suffered a secular decline in financial strength in the 80 years since the Federal Reserve System was established in 1913. For example, capital ratios have fallen from 20 percent at the turn of the century to around 6 percent today. Banks are also far less liquid today than they were in earlier decades. The loan quality of banks has declined steadily over our central banking era. Profitability has been weak and irregular compared to the period before central banking. Finally, bank failures have been more a problem under central banking than under previous banking eras in U.S. history.[4]

To be sure, these measures of banking system strength have ebbed and flowed cyclically over the past eight decades—for example, the dissolution of the 1930s, the seeming calm of the 1950s, and the renewed turbulence of the past two decades. But in my own work, I’ve identified an undeniably pronounced secular decline in the financial condition of banks, in good times and bad. This leads me to question the legitimacy of central banking as such. I’m encouraged to find that other scholars are also questioning the conventional wisdom about central banking.[5]

I attribute the secular decline of banks to central banking not only because that has been the predominant structure governing our money and banking system for most of this century, but because the main features of central banking bear directly on the worsening finances of the banks.

For example, central banking involves a legal tender monopoly on the production of paper currency, and to the extent this money is produced in excessive supply and forms the base of banking system deposit expansion, it inflates bank balance sheets and invites malinvestment of resources. Central banking is characterized by a lender of last resort function that can be seriously mismanaged, as it was in the 1930s, causing widespread bank failures. Central banking is usually accompanied by a system of flat-rate federal deposit insurance, a system known by all to promote excessive risk-taking and imprudence among banks.

It should not have taken decades to see this would happen. Back in 1908, when earlier versions of government deposit insurance were advanced, the president of the First National Bank of Chicago, James Forgan, asked the following: “Is there anything in the relations between banks and their customers to justify the proposition that in the banking business the good should be taxed for the bad; ability taxed to pay for incompetency; honesty taxed to pay for dishonesty; experience and training taxed to pay for the errors of inexperience and lack of training; and knowledge taxed to pay for the mistakes of ignorance?”[6]

As I have argued elsewhere, “deposit insurance is a scheme put in place because the Federal Reserve mismanaged the discount window in the 1930s, and it is a scheme that has been expanded ever since in concert with the Fed’s inflation of the money supply (which consists predominantly of bank demand deposits).”[7]

Finally, systems of central banking involve extensive regulation of bank branching, lending, and product offerings—regulations that prohibit sound diversification and invite still greater instability.

Unsafe and Unsound

If the purpose of central banking is to ensure sound money and safe banking, then central banking has been an unmitigated failure. I have already summarized the relative decline of banking’s strength as captured in financial ratios. But the purchasing power of money has also declined, so that a 1913 dollar is worth ten times more than a 1992 dollar. We enjoyed much sounder money and safer banking in the eight decades before central banking was established here in 1913 than we have in the eight decades since. I conclude that this is so because central banking represents a special case of the general failure of central economic planning, a failure that most of the world is only now beginning to recognize.[8]

The fact that central banking flies in the face of free-market alternatives is recognized by some of its most prominent practitioners. In a symposium sponsored by the Federal Reserve Bank of Kansas City in August 1990, Paul Volcker noted, “Central banks were not at the cutting edge of a market economy . . . . Central banking is almost entirely a phenomenon of the 20th century . . . . Central banks were looked upon and created as a means of financing the government. . . . If you say central banking is essential to a free market economy, I have to ask you about Hong Kong, which has no central bank at all in the absolute epitome of a free market economy. Yet it does quite well in terms of economic growth and stability.”[9]

My research confirms Mr. Volcker’s assessment. The primary purpose of central banking is to finance the government.[10] That’s what it does consistently and what it does best—and does so, unfortunately, at the expense of sound money and safe banking. Mr. Volcker would find results in the U.S. similar to those of Hong Kong, as I did, by examining the decades before the Federal Reserve was established.

In the eight decades before 1913 we had a system which can very loosely be called “free banking and the gold standard.” There was no central bank, no lender of last resort, no federal deposit insurance. Banks issued currency as well as checking deposits, convertible into the precious metals. Bank note redemptions and the gold standard anchored the money supply. Excessive currency issuance was prevented. Money expanded and contracted with the needs of trade, not with the needs of government. Banks formed clearing-houses to settle balances and they lent on an inter-bank basis to temporarily illiquid but solvent institutions. The few banks that failed were absorbed into stronger ones or simply liquidated at a discount to noteholders.[11]

The free banking era was not totally free, of course. Bank note issues were restricted by laws requiring currency to be backed by state or federal bonds—an indirect means of financing government. Branching was restricted as well, preventing full diversification. But the U.S. free banking era was more in line with a free market system of money and banking than our present era. As such, it should not be surprising that it produced relatively higher quality money and much safer banking. I document these facts in my book. For more background on the favorable history of the free banking era, I recommend the work of Arthur Rolnick and Warren Weber at the Federal Reserve Bank of Minneapolis.[12]

Only with this wider historical and theoretical context can we grasp the full implications of today’s “too-big- to-fail” doctrine. In my view, banking without the “too-big-to-fail” doctrine is not simply banking prior to 1984, the year when Todd Conover, Comptroller of the Currency, said the top 11 banks in the country would not be permitted to fail. For me, banking without “too-big-to-fail” is banking before 1913, the year when the Federal Reserve was established. For as I have indicated, the doctrine is inextricably linked with central banking. No free market system of money and banking would aim to sustain insolvent institutions, and there would be no institutional bias in favor of generating insolvent institutions, as central banking engenders. Free banking minimizes the spread of problem banks from the very start. No central bank monetary inflation or taxpayer deposit guarantees are employed to force-feed a free banking system.

Undermining the Financial Integrity of Banks

In two important respects, the “too-big-to-fail” doctrine represents an unhealthy extension of two central banking features that have already been shown to undermine the financial integrity of banks.

First, the “too-big-to-fail” doctrine has transformed the lender of last resort from one providing cash to temporarily illiquid banks to one providing extended credit to permanently insolvent banks. One of the first theorists of the lender of last resort function, Walter Bagehot, warned us that there would be times when a central bank couldn’t effectively distinguish between illiquidity and insolvency.[13] But in recent years the discount window has been thrown wide open to banks widely admitted to be insolvent. For example, a 1991 House Banking Committee report concluded that the central bank provided subsidized credit to hundreds of banks that ultimately failed. In six years ending May 1991, 530 of the 3,000 banks that drew on the discount window failed within three years. Many more, if not outright failures, had the lowest financial performance ratings assigned by regulators.

Even as a provider of short-term liquidity, the lender of last resort offers a safety valve for banks that do not properly manage their liquidity positions. This subsidy for liquidity mismanagement has been in place for years. We were always assured that the Fed would manage access to the window with prudence and discretion. But now this malincentive has been extended still further to cover up the insolvency of banks. Perhaps even worse, access to the discount window was widened in the 1991 banking law to include the securities industry. More recently, there was talk among U.S., British, and Canadian central bankers of assisting real estate developer Olympia and York, on the grounds that its bad loans would harm big banks. There appears to be no end to the degeneration of the lender of last resort function.

Second, the “too-big-to-fail” doctrine has unwisely extended deposit insurance coverage from insured depositors to uninsured depositors and creditors. Government guarantees of insured deposits are bad enough in the way they promote reckless banking. The more than doubling of deposit coverage in 1980 institutionalized the recklessness. The extension of coverage to all creditors of banks, as under the “too-big-to-fail” doctrine, is the height of irresponsibility.

Nothing in the 1991 banking law removes the discretion of the Fed or the Treasury in employing “too-big-to-fail” at any time for any purpose.[14] To the extent the doctrine has not been employed as extensively in recent failures, it seems only because of the insolvency of the deposit insurance funds themselves. “Too-big-to-fail” is not a doctrine which can be effectively scaled back in isolation or in increments. Unless there is an outright rule against it, exceptions will always be made to expand it.

Bad as they already were, discount window activity and deposit insurance coverage have degenerated further in recent decades, in the name of the “too-big-to-fail” doctrine. We need to repeal the structural central banking features that generate failed banks, not simply patch on some extended version of these features, a patch job supposedly justified by pointing to all the failures. Accompanying the unconditional repeal of “too-big-to-fail” must be a scaling back and eventual abolition of federal deposit insurance and discount window lending as well. The sooner this occurs, the sooner banking will be restored to the health it enjoyed before these features were in place.[15]

The Fear of Contagious Bank Runs

Opponents of the repeal of the “too-big-to-fail” doctrine often cite the so-called “contagion” effect of bank failures, the domino effect of large bank failures precipitating other failures, allegedly cascading into a system-wide collapse.

In my estimation, no factor contributes more to this risk than government restrictions on branching. U.S. banking historians know all too well that widespread correspondent banking and extensive reliance on inter-bank deposits in this country stem directly from branching prohibitions.[16] In nationwide banking systems, such as in Canada, inter-bank exposures are minimal.[17] But in the U.S., the government has promoted an interlocking banking system, in effect requiring banks to line up like dominos, preventing them from holding their own direct deposits in their own chosen areas of the country. Having created such unstable links, government has then advanced a “too-big-to-fail” doctrine to prevent smaller banks from being harmed by losses on deposits at bigger banks.

Here is an obvious case of government interventions that have bred further intervention, allegedly to remedy the distortions brought about by still earlier interventions. Eugene White and others have shown that U.S. banking history is replete with evidence of this vicious circle.[18] There is only one solution to this madness, and that is to repeal the interventions across the board. Let’s start by permitting what every advanced country permits of its own banks—the ability to branch freely and diversify their operations.

I will not repeat here in detail other important refutations of the so-called “contagion” argument, especially those made by economist George Kauffman.[19] Suffice it to say, he argues that if some banks are weak, depositors will transfer their money to stronger ones. If they don’t find stronger ones they will make a flight to quality and acquire government securities, the sellers of which must be confident of finding stronger banks, because in selling they expect to deposit the cash proceeds. In either of these cases, there is a redistribution of reserves, but no destruction of them. There is no deflation of the aggregate money supply and hence no contagion effect.

What if the strength of all banks is doubted by all parties? Then there will be a flight out of deposits into currency, a precipitous drop in the deposit/currency ratio so common to deflations. A loss of reserves could kick off a multiple contraction process that affects healthy banks as well as insolvent ones. But observe that such deflations are exacerbated by fractional reserve banking, and especially by very low fractions. Economists who recognize this potential problem tend to argue for some form of deposit insurance to contain it. I believe, to the contrary, that all government deposit insurance is destabilizing. I oppose it on principle, mindful of the fact that even limited forms of it soon grow into uncontrollable excess.

Furthermore, my own research indicates that bank liquidity is far lower—that is, reserve fractions are far lower—under central banking than under free banking. Hence a deposit contraction is potentially more severe when a central bank is in charge. More important, free banking offers a direct solution to the problem. A system of free banking permits private bank currency issuance, so banks can easily meet shifts in customer demand for currency relative to checking deposits. Such shifts are far less easily accommodated by a monopoly currency issuer which can misjudge and mismanage the shift, as did the Federal Reserve in the early 1930s.

On these grounds alone, I believe there is good reason to secure some end as well to the legal tender laws which grant a monopoly on currency issuance to the Federal Reserve. I have offered other reasons for the repeal of the legal tender laws in my book. As the late Nobel Prize-winning economist Friedrich Hayek argued, we need “a denationalization of money,” and the kind of choice in currencies that brought us stable money and banking in the 19th century.[20] Parting somewhat from Hayek, I believe this free issuance of bank notes must also involve gold-convertibility, as note issue did during our better banking era.

A proper legal structure upholding property rights is also important. Free banking does not entail anarchy. Contracts must be enforced. The repeal of the “too-big-to-fail” doctrine will not be truly sustainable unless banks are fully subject to the general bankruptcy laws. No other industry is exempt from such laws, nor so harmed by the exemption.

Until and unless banks are subject to bankruptcy, we will continue to see failures handled according to politics and bureaucratic motives—such as agency “image”—not according to simple justice and sound economics. We will continue to witness swings from a regulatory policy of “forbearance” to a policy of “early intervention,” to forbearance, and back again. Both policies are detrimental to the banking system, and not only because of their unpredictable application from one case to the next or one year to the next.

Forbearance, as is known to all, promotes laxity in accounting and financial control, condoning, if not encouraging, recklessness, hiding insolvency, and ballooning ultimate losses. “Early intervention,” on the other hand, has its own dangers. While posing as a remedy for the ills of forbearance, a policy of early intervention actually holds out the very definite prospect of de facto nationalizations of the banks. After all, if banks with 2 percent capital ratios are to be closed down or taken over, as provided in the 1991 banking law, what else can such a policy be called but a nationalization, indeed a “taking,” under the Fifth Amendment? The recent nationalization of Crossland Savings Bank offers a chilling precedent for this disturbing new extension of the “too-big-to-fail” doctrine.[21]

If, instead, banks are subject to the bankruptcy laws, the competing interests of management and creditors, including the creditors who are depositors, will prevail. Closures of failed institutions will not be sudden but orderly. They’ll be drawn out in a rational manner, but not forever, as in the case of the thrifts or the Rhode Island credit unions. Neither will closures under bankruptcy take place prematurely, while there remains value in the franchise. For a more detailed look at this approach, I commend to you the work of Robert Hetzel at the Federal Reserve Bank of Richmond.[22]

In conclusion, I want to stress that the “too-big-to-fail” doctrine is part and parcel of a wider system of central banking that undermines the financial condition of the banking system. The sooner we phase out this system in favor of free banking and the rule of law, the better off we will be. In other words, repealing the “too-big-to-fail” doctrine will be a good start, but it won’t go far enough in curing what really ails the banks.

1.   Irvine H. Sprague, Bailout.’ An Insider’s Account of Bank Failures and Rescues (New York: Basic Books, 1986).
2.   Paul A. Samuelson and Herman E. Krooss, Documentary History of Banking and Currency in the United States, Volume IV (New York: Chelsea House Publishers, 1983), p. 354.
3.   Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867-1960 (Princeton, N.J.: Princeton University Press, 1963), Chapter 7.
4.   Richard M. Salsman, Breaking the Banks: Central Banking Problems and Free Banking Solutions (Great Barrington, Mass.: American Institute for Economic Research, 1990).
5.   See especially Lawrence H. White’s works, Free Banking In Britain: Theory, Experience and Debate, 1800-45 (Cambridge: Cambridge University Press, 1984) and Competition and Currency: Essays on Free Banking and Money (New York: New York University Press, 1989).
6.   lames B. Forgan, “Should National Bank Deposits Be Guaranteed by the Government?” Address to the Illinois Bankers’ Association, June 11,1908 (Chicago: First National Bank of Chicago).
7.   Richard M. Salsman, The Credit Crunch: Myth or Reality? American Institute for Economic Research, October 1991.
8.   Economists of the Austrian School of economies, especially Ludwig von Mises and Friedrich Hayek, have been identifying this failure for most of this century.
9.   Paul Volcker, “The Role of Central Banks” in Central Banking Issues in Emerging Market. Oriented Economics (a symposium sponsored by the Federal Reserve Bank of Kansas City, August 23-25, 1990). Definitive historical evidence for Volcker’s summary assessment can be found in Charles Goodhart’s The Evolution of Central Banks (Cambridge: The MIT Press, 1988).
10.   Salsman, Breaking the Banks, chapter 8.
11.   Salsman. ibid., chapter 6.
12.   See especially Arthur J. Rolnick and Warren E. Weber, “Free Banking, Wildcat Banking, and Shinplasters.” Quarterly Review, Federal Reserve Bank of Minneapolis, Fall 1982, pp. 10-19.
13.   Walter Bagehot, Lombard Street: A Description of the Money Market (London: Kegan, Paul & Co., 1873).
14.   See the misnamed Federal Deposit Insurance Corporation Improvement Act of 1991 (FIDICIA).
15.   I have explained in detail how this might be accomplished in Chapter 9 of Breaking the Banks.
16.   Walker Todd and James Thompson, “An Insider’s View of the Political Economy of the ‘Too-Big-To- Fail’ Doctrine.” Federal Reserve Bank of Cleveland, Working Paper #9017, December 1990. p. 16.
17.   Lawrence Kryzanowski and Gordon Roberts, “The Performance of the Canadian Banking System, 1920- 1940,” Proceedings from a Conference on Bank Structure and Competition (Chicago: Federal Reserve Bank of Chicago, May 1989). pp. 221-232.
18.   Eugene Nelson White, The Regulation and Reform of the American Banking System, 1900-1929 (Princeton, N.J.: Princeton University Press, 1983).
19.   George Kauffman, “Are Some Banks Too Large to Fail?” Federal Reserve Bank of Chicago Working Paper, June 1989.
20.   Friedrich A. Hayek, Denationalization of Money (London: The Institute for Economic Affairs, 2rid Edition, 1978).
21.   Jonathan R. Macey, “Needless Nationalization at the FDIC,” The Wall Street Journal, February 14,1992. According to Mace),, “By nationalizing Crossland, the FDIC is signaling that it can take over any bank or thrift it wants, no matter how large or small, or how remote the threat to the banking system.”
22.    Robert Hetzel, “Too Big To Fail: Origins. Consequences, and Outlook,” Economic Review, Federal Reserve Bank of Richmond. November/December 1991, pp. 3-15.