A perplexed economist confronts 'too big to fail'.

• Author: Scherer, F.M.
• Position: Report

1. Introduction

   The bailouts rescuing failing financial institutions that both the U.S. government and European governments have found themselves compelled to sustain during the recent economic crisis have repeatedly raised the question, "How big is too big?" in bank size. For me, as a more or less traditional industrial organization economist, this poses a particular challenge. From early in my career, I had drummed into my head the mantra of MIT's Morris Adelman, "Absolute size is absolutely irrelevant." (3) And what we appear to be addressing is in fact absolute financial institution size.

   What Adelman was saying is that the core understanding of industrial organization economists, built up over decades of research, was that monopolistic influences on price-setting depended upon relative size--i.e., the size of the price setters relative to the market in which they operated. Pure monopoly is of course the extreme case: the monopolist commands close to 100 percent of the relevant market. Oligopoly is the case that has proved both most interesting in the real world and most difficult. The individual sellers are large relative to the market, and a few of them have sufficiently large market share, so that they are acutely aware of their interdependence in pricing. Given this size relationship, as economists from Augustin Cournot (1838) to Edward Chamberlin (1933) insisted, individual firm sizes sufficient to give the oligopolistic sellers a large share of the markets they serve were likely to facilitate prices elevated about the competitive level, inefficient resource allocation, and maldistribution of income. But this is not my initial 'too big' focus. I shall return later to argue that it may be more relevant than we might casually assume.

   Although Morry Adelman rarely missed the mark, in this case, I believe, absolute size does have relevance. There are at least three reasons why this may be true.

   First, as every newspaper reader recognizes, financial institutions may be so large relative to the whole world of finance, regardless of their size in individual price-setting contexts, and also so interdependent in their relationships, that failure of one or more institutions has systemic consequences. That is, one large institution's failure, and its attendant inability to meet obligations to a host of financial counter-parties, can jeopardize the health of numerous other banks. And if many banks' credit "freezes up" as a result of these failures, the prosperity of the economy as a whole can be jeopardized. This is a real and serious problem, one known to economists since at least the time of Adam Smith.

   Second, absolute size may carry not only this hazard, but it may also yield economies of scale and scope that make individual banks better able to perform their vital functions, providing credit to economic actors on more favorable terms. Whether this is true is an empirical question. I shall return to it, constrained to be sure by severe limits on economists' knowledge of such matters.

   Third, financial institutions that are large in absolute size may have deep and well-filled pockets with which they can among other things hire lobbyists, support individual political parties and election candidates, and, under the recent Supreme Court reinterpretation of the U.S. Constitution's first amendment, mount advertising campaigns in direct support of or opposition to election candidates. (4) In Federalist Paper No. 10, James Madison warned against the political power of factions resulting from "the verious and unequal distribution of property" ... " who are united and actuated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community." Robert Bork (1966; 1978, Chapter 2) has argued that concern over the resource misallocation that comes from monopoly power in specific markets--i.e., relative size--was what primarily motivated the U.S. Congress to enact the Sherman Antitrust Act in 1890. Critics such as Robert Lande and I have argued that Bork's interpretation is erroneous. (5) I leave for another forum the more detailed exploration of this debate.

   Here I offer only one additional strand of historical evidence--Figure 1, drawn by artist Joseph Keppler for Puck magazine, January 23, 1889--a year before the Sherman Act was passed. One cannot view it without recognizing that the U.S. public was alarmed at the time about the political power of the great trusts, for which we might now substitute bloated figures for JP Morgan Chase, Bank of America, and Goldman Sachs. It would be hard to deny that public concerns over the trusts' power in federal and state legislatures were an important stimulus to the Sherman Act's passage. (6) And now, 120 years later, there is abundant reason to fear the enormous political power of the financial institutions, said by Drum (2010) to own Washington "lock, stock, and barrel." In 2008, for example, the finance lobby is said to have contributed $475 million to political candidates and their supporting party organizations--more than twice the level of contributions from the second-largest lobby, the health care industry. (7) To be sure, an industry of relatively small entities such as farmers ($65 million in 2008) might amass large political contributions through the efforts of industry-spanning trade associations. But what evidence we have suggests that collective action groups such as the American Bankers Association were relatively minor factors in the torrent of political donations. (8)

2. The Stylized Facts

   Let me proceed by laying out what economists call "stylized facts"--that is, parcels of evidence without direct theoretical or proven causal connections to the issues of bank size. I then proceed to examine what we actually know about causal links.

   The first salient fact is that the banking industry has experienced during the past three decades a merger wave of monumental proportions. Dean Amel of the Federal Reserve Board staff reports (2002) that between 1990 and 2001--i.e., before the mega-mergers precipitated by the 2008 financial crisis--U.S. banks consummated mergers and acquisitions valued at more than $900 billion. (9) My own attempt to determine what happened is best reported in two steps.

   First, from Fortune magazine's annual lists of the 100 largest commercial banking and diversified financial companies, I began with the listing published (from 1984 financial reports) of the largest corporations as of 1985 and traced what happened to them by the close of 2008. Among the 30 leaders ranked by assets as of 1985, only nine survived in more or less recognizable form at the end of 2008. Eighteen of the 30 disappeared through mergers; three failed and were liquidated by governmental financial guarantors.

   The second step is embodied in Figure 2. It traces principal events in the merger histories of six corporations that by the end of 2008 had become the largest U.S.-based financial entities measured by asset volume. (10) Altogether, 53 substantial components are found to come together into six surviving entities, ranked in order of end-of-2008 assets. The 1985 asset ranks of the merging entities are given in parentheses following the company names. Not all of the named survivors were the first movers in mergers that led to substantial consolidation. In four cases marked [circle L], another bank took the lead, choosing after consummating a merger to adopt a new name based upon the name of its acquisition target. The analysis was able to track only the most significant mergers. At the end of each surviving company trajectory is a number followed by "SM," for small mergers. That number was obtained by tracking smaller acquisitions reported in the company histories published in Moody's (now Metgent's) Bank & Finance Manual. It is probably incomplete, but altogether, 139 acquisitions too small to be accommodated in Figure 2 were tabulated. (11) Clearly, the industry's merger-based structural transformation has been profound.

   A related stylized fact has been the increasing concentration of total U.S. financial assets held by the largest institutions. It is shown by Figure 3, drawn without change from page 100 of an excellent book by Henry Kaufman (2009), former managing director of Salomon Brothers. Until the end of the 1980s decade, there was a gradual decrease in the concentration of financial institution asset holdings. After that, a striking upsurge occurred. In 1990, the largest ten financial companies controlled a bit less than 10 percent of total U.S. financial institution assets; by 2004, when Kaufman's series ends, their share exceeded 50 percent. The top 20 institutions controlled 14 percent of assets in 1990, rising to 63 percent in 2004. (12) After 2004 the concentration process undoubtedly continued as the largest institutions absorbed huge financial intermediaries brought into jeopardy by the crisis of 2008.

   The third key stylized fact is presented in Figure 4. (13) It shows for 1960 through the third quarter of 2008 profits (before income taxes) reported by U.S. financial corporations as a percentage of total domestic industries' corporate profits. The financial sector's share fluctuated in the range of 8 to 18 percent up to the late 1980s, after which a sharp...