Sowing the Seeds of a Future Crisis: The SEC and the Emergence of the Nationally Recognized Statistical Rating Organization (NRSRO) Category, 1971–1975

Abstract

Since 2008, academics and policymakers have frequently debated why bond rating agencies such as Moody's, S&P, and Fitch enjoy considerable power and influence. The 2008 financial crisis focused our attention on the bond rating agencies that had previously categorized mortgage-backed securities as investment grade. Scholars have attributed the power enjoyed by the rating agencies to regulations that confer a privileged status on those agencies that are designated as nationally recognized statistical rating organizations (NRSROs) by the U.S. Securities and Exchange Commission (SEC). While these authors mention in passing that the relevant regulation went into effect in 1975, none has conducted archival research to examine why this regulation was introduced at that time. This article is the first historical investigation of the creation of this crucial regulation, which entrenched the concept of the NRSRO in federal securities law. It shows that the SEC mandated the use of NRSRO-created ratings even though SEC officials vigorously debated whether it was wise for the commission to endorse ratings produced by agencies that operate on the basis of the controversial issuer-pay model. This article contributes to our understanding of the SEC's role in the development of the distinctive features of American capitalism.

The global financial crisis (GFC) of 2007 to 2009 and the subsequent Eurozone crisis focused the attention of researchers on the bond rating agencies that had previously categorized mortgage-backed securities as investment grade.¹ One strand of this research seeks to explain the enduring influence of these agencies in global securities markets despite the accumulation of evidence that their ratings have poor predictive power. Researchers who use the concept of “regulatory license” have argued that the persistent power of the bond rating agencies is a result of regulations that induce financial institutions to hold securities that are highly rated by those credit rating agencies that the U.S. Securities and Exchange Commission (SEC) has designated as Nationally Recognized Statistical Rating Organizations (NRSROs). Ideologically diverse academics have discussed the power NRSROs enjoy, as have political leaders concerned about the erosion of national sovereignty and the cost to the taxpayer of bank bailouts. Writing in the aftermath of Greece's debt crisis, one scholar condemned Moody's and S&P as an “unelected cabal of private agencies.”²

Those who have written about the regulatory license possessed by the NRSROs have mentioned, in passing, that the relevant regulation went into effect in 1975 via a modification of SEC Rule 15c3-1, which was changed to encourage broker-dealers to hold securities that had been deemed investment grade by “nationally recognized” bond rating agencies. An SEC “no-action letter” published in March 1976 clarified that when the SEC referred to nationally recognized ratings agencies, it meant Moody's, S&P, and Fitch—the so-called Big Three.³ Economists, international political economists, and legal scholars have shown that referencing NRSROs in the 1975 net capital rule was highly consequential.⁴ Finance academics have demonstrated that when a rating agency is granted NRSRO status by the SEC, the agency swiftly acquires the power to change the capital costs of the firms it rates.⁵ The commission into the causes of the crisis that reported to the president in 2011 even described the 1975 rule change as a “watershed event” because of the power thus granted to the NRSROs.⁶ However, no scholar has closely examined why the rule was introduced. This article is the first historical investigation of the creation of this crucial regulation, which entrenched the concept of the NRSRO in federal securities law. It shows that the SEC mandated the use of NRSRO-created ratings even though its staffers privately expressed serious concerns about the reliability of those ratings and advocated the use of mark-to-market (MTM) approaches in interpreting the net capital rule.

Given the globally controversial power of U.S.-based bond rating agencies, historical research on their origins remains topical. Moreover, many historians have in the last decade come to view the 1970s as a crucial turning point in the history of American economic policy that witnessed changes in the nature of government regulation, the onset of a period of falling growth and increased inequality, and the financialization of the economy.⁷ Historians who study other advanced economies also regard the 1970s as an era in which financialization accelerated.⁸ Financialization involves an increase in the proportion of economic activity, political power, and workers concentrated in the financial sector.⁹ While some authors argue that deregulation in the 1970s and 1980s caused subsequent financialization,¹⁰ economist John Kay cautions that changes in regulatory detail, and not deregulation per se, explain the growing importance of finance, particularly financial activities of limited social value. He argues that the decades immediately before the 2008 financial crisis witnessed an unfortunate shift in the focus of regulators away “from protection of consumers to protection of markets” owing to the political power of the incumbent firms that influence regulatory design.¹¹ His insight informs our research into the making of the NRSRO category. Our research is also grounded in the historiography suggesting that while regulatory changes in the 1970s alleviated some pressing challenges, they also sowed the seeds of other problems that became manifest in the twenty-first century. We suggest that the NRSRO mandate embedded in the 1975 net capital rule should be added to the list of 1970s policy reforms that appeared to solve an immediate problem while creating trouble in the future.

The SEC and the Bond Rating Agencies: Theoretical Lenses

In the 1960s, activists such as Ralph Nader and Wilma Soss highlighted the failure of regulatory agencies including the SEC, the Federal Communications Commission (FCC), and the Food and Drug Administration to protect consumers and other stakeholders from corporate malfeasance.¹² Nader's ideas anticipated the work of academics from both ends of the political spectrum who advanced theories that critiqued existing scholarship on regulations for naïvely accepting agency claims that their regulations were motivated by genuine concern for the public interest.¹³ The basic insight is that regulatory agencies are staffed by self-interested individuals who sometimes exchange favors with the firms that they regulate. The new literature used terms such as “the revolving door” and “regulatory capture” to explain how federal agencies made policy that benefited incumbent firms at the expense of the public.¹⁴ Thomas Hazlett's recent history of the FCC illustrates how this perspective can enhance historical research on regulatory agencies.¹⁵

One approach that others have used to view decision-making at regulatory agencies such as the SEC is the bureaucratic imperialism theory, which conceptualizes regulatory agency employees as interested in expanding the power of “their” agency. According to this theory, officials engage in bureaucratic turf wars with rival regulatory agencies and work tirelessly to expand their budget and power. SEC officials obtain “career support” by working with interest groups that benefit from increased regulation, including lawyers and the securities firms with the greatest capacity to absorb compliance costs. In the securities industry, as in other industries, incumbent firms can support the introduction of regulations that raise rivals’ costs.¹⁶ The bureaucratic imperialism theory also helps one to explain the SEC's well-documented tendency to make policy through no-action letters rather than through publication of clear rules. Ambiguous regulations increase SEC staff member discretion and power.¹⁷ The theory also explains why the number and complexity of SEC regulations increased steadily over time.¹⁸

Bureaucratic imperialism theory inadequately accounts for instances where the SEC deliberately withdrew from regulatory activities, as when it permitted the growth of a large unregulated private securities market after the GFC. To remedy that oversight, legal scholar Zachary Gubler introduced the concept of “political slack,” a measure of public scrutiny of an agency's activities.¹⁹ When political slack is low, agency staff typically refrain from expansionist behavior for fear of attracting unwanted attention. When political slack is high, agencies are more likely to behave imperialistically. Gubler's concept helps to explain the actions of the SEC in the 1970s, as the relevant changes of regulation received scant public attention prior to the financial collapse of New York City in late 1975, months after the SEC finalized its new net capital rule in June.

The NRSRO Category

Before the GFC, approximately 150 credit rating agencies existed but the SEC had designated only 10 as NRSROs, a coveted status because several regulatory agencies require that firms consult NRSRO ratings when making investment decisions. A 2003 internal report observed that after the SEC enshrined NRSRO status in 1975, the term began to appear in a wide variety of federal, state, and non-American regulations, including those that apply to universities and insurance companies.²⁰ Today, the Big Three NRSROs, Standard & Poor's, Moody's, and Fitch, control roughly 85 percent of the securities rating market. Only two small NRSROs are headquartered outside of the United States, a fact stressed by scholars and political leaders who see NRSROs as instruments of American imperialism.²¹

The collapse of Enron in 2001 focused attention on the bond rating agencies that had previously endorsed its securities as investment grade. During congressional hearings into the agencies in 2002, critics complained that the SEC's criteria for designating bond rating agencies as NRSROs had never been codified and published. During those hearings, witnesses spoke of “very high barriers to entry with respect to the NRSRO designation” and legislators considered what would happen if the SEC “allowed credit rating agencies to exist on a free-market basis,” by which they apparently meant the elimination of privileged NRSRO status.²² The Credit Rating Agency Reform Act of 2006 required the SEC to publish, for the first time, its criteria for determining which bond rating agencies qualified as NRSROs. The GFC intensified demands for reform and in 2010 Congress required the SEC to create an Office of Credit Ratings to enhance the regulation of NRSROs.²³

Historiography of the Bond Rating Agencies

Although scholars have studied the early history of the broader credit-rating industry, the historiography of bond rating in the second half of the twentieth century remains sparse.²⁴ Sarah Quinn briefly discusses Moody's in her history of American bonds but focuses only on its origins in the Gilded Age, when investors were confronted with a rapidly growing number of securities issued by distant firms, not on its postwar growth.²⁵ A 1974 Century Fund think tank report includes a historical overview on municipal bond rating. This report was commissioned by the Century Fund trustees, who included Federal Reserve chairman Arthur Burns, to address growing concerns that the agencies wielded excessive power over municipal borrowing costs and issued subjective ratings. Municipal finance expert John E. Petersen, the report's primary author, traces bond rating to John Moody, who began publishing bond ratings in 1909 using alphabetical codes like “AAA” to indicate likelihood of default. These ratings simplified decision-making by investors operating in the increasingly complex and opaque capital markets of that era. S&P imitated Moody's, and both firms later expanded into rating state, municipal, and foreign government bonds.

In his report, Petersen observes that the total workforce of Moody's had not exceeded four people between 1920 and 1935 and that the acquisition by Moody's and S&P of large numbers of professional analysts was a new development that had been made possible, in part, by the agencies’ recent shifts to the issuer-pays business model.²⁶ Petersen suggests that the 1938 interagency agreement to require banks to consult ratings manuals—an agreement made by the Comptroller of the Currency, the Board of Governors of the Federal Reserve, the Federal Deposit Insurance Corporation (FDIC), and the National Association of Supervisors of State Banks—had amplified the importance of ratings. He also suggests that Moody's, the older of the two major agencies, retained the benefits of first-mover advantage because Moody's issued more ratings, though S&P gave issuers somewhat higher ratings. Peterson reports that circa 1970, 60 percent of the dollar volume of bonds were rated by both agencies, 35 percent were rated only by Moody's, and 5 percent were rated only by S&P.²⁷

In the first major scholarly work on the history of the bond rating agencies, Richard Sylla argues that such agencies emerged in the United States in the early twentieth century in response to the increasing size and complexity of the U.S. corporate bond market, which made it difficult for investors to determine issuer creditworthiness. Sylla describes how bond ratings emerged in the early twentieth century in response to the needs of investors operating in an increasingly complex environment in which it was difficult to ascertain the creditworthiness of distant bond issuers, and he describes how Standard Statistics and Fitch imitated Moody's innovative ratings. He suggests that the agencies’ business model shifted in the 1970s, as the bond rating agencies ceased to derive most of their revenue from selling reports to investors and instead came to rely on fees paid by the issuers. Sylla observes that the workforce of the leading ratings agencies expanded dramatically following the shift to the issuer-pay model. Whereas previously the agencies had just a “handful” of analysts, by 1995 S&P employed 800 analysts and Moody's 560. Many critics of the Big Three focus on the conflicts of interests associated with this business model, but Sylla suggests that this “cynical” view is too harsh and overlooks the tremendous value that the agencies’ hardworking analysts have created. He suggests that the decision of the SEC in the 1970s to designate specific agencies as NRSROs was a minor detail rather than a decisive turning point.²⁸

In contrast, Frank Partnoy stresses the importance of the 1975 regulatory change, arguing that the regulatory license the SEC granted effectively cartelized the bond rating industry by creating regulatory barriers to entry. The informational value of the ratings could therefore plummet, and indeed did, leading to multiple crises culminating in the GFC. Partnoy attributes the usually high profit margins of the NRSROs to regulation.²⁹ Economist Lawrence J. White observes that in 1975 the SEC followed a precedent set when the Comptroller of the Currency began outsourcing decisions about the prudence of bank investment practices to the bond rating agencies. The Comptroller did so by requiring commercial banks to invest only in those securities that were listed as investment grade in “recognized rating manuals.” This term, which appeared in the text of a 1936 regulation, was likely a reference to the manuals sold by the Big Three.³⁰ According to White, the 1975 SEC regulation extended a rule that the government had previously applied to banks to cover the nation's broker-dealers, a category of firms that included major investment banks, retail brokerages, and securities houses.³¹

In explaining why the SEC introduced this regulation in 1975, White observes that the surprising 1970 bankruptcy of Penn Central ended a long period during which few major American corporations had defaulted. A wave of broker-dealer failures had further highlighted the importance of risk-weighting broker-dealer asset portfolios. He speculates that the SEC officials responsible for the 1975 regulation must have been worried that a new bond rating agency might appear that would certify low-quality securities as investment grade. To prevent the rise of such a bogus rating agency, the SEC modified Rule 15c3-1 in 1975 so that broker-dealers could only use ratings produced by “nationally recognized” bond rating agencies.³² White's account implies that this feature of the 1975 net capital rule was, in retrospect, a blunder that gave the Big Three a regulatory license they subsequently abused. White's account of the SEC's rationale for mandating the use of bond ratings is based on neither archival research nor oral history.

In their analysis of why U.S. financial regulators empowered specific rating agencies, economic historians Marc Flandreau and Joanna Kinga Sławatyniec stress institutional path-dependence and agree with Partnoy's view that American regulators’ habit of outsourcing decisions to bond rating agencies is problematic, arguing that the SEC and other U.S. regulatory agencies are “addicted” to the ratings produced by the Big Three. They trace this addiction to a series of court cases in the early twentieth century that had legitimated the idea that the use of bond ratings was a sign of good business judgment. Entrenching the use of Big Three ratings in regulation, Flandreau and Sławatyniec argue, merely followed judicial precedent.³³ Although their research on the nineteenth and early twentieth centuries deepens our understanding of the more distant roots of SEC Rule 15c3-1, it does not help to explain why the SEC introduced this rule in 1975.

The SEC's Profile

Section 4 of the Securities Exchange Act of 1934 created the SEC in the wake of the Pecora Hearings, which had convinced many Americans that existing securities regulations, including state “Blue Sky” laws and the self-governance systems of the New York Stock Exchange (NYSE) and other exchanges, insufficiently protected retail investors. The statutory goals of the SEC included protection of investors, creation of fair and orderly securities markets, and encouragement of capital formation.³⁴ Most of the investment industry initially opposed the SEC's creation, as did conservatives who disliked New Deal agencies. Under the SEC's first chairman, Joseph P. Kennedy, however, most securities market professionals concluded that the SEC's existence was compatible with their interests. Thereafter, the SEC's existence ceased causing substantial political controversy. Although Richard Nixon criticized the SEC for unspecified “heavy-handed bureaucratic regulatory schemes” during his 1968 presidential campaign, he did not curtail its power after his election. Subsequent Republican presidents also accepted the SEC's legitimacy, even while attacking or proposing the abolition of other federal regulatory agencies.³⁵

Five presidentially appointed commissioners supervise the SEC. All serve five-year terms staggered so that one commissioner's term expires each year. No more than three commissioners may come from the same political party. In 1970, the SEC had three main divisions: Corporate Finance; Corporate Regulation; and Trading and Markets. The SEC also included the office of General Counsel, nine regional offices, and seven branch offices in important cities.³⁶ The SEC's Trading and Markets Division appears to have had the greatest role in the design of the 1975 uniform net capital requirement and thus the NRSRO designation. It oversaw broker-dealer firms, investment banks, and self-regulatory organizations such as the Financial Industry Regulatory Authority (FINRA) and Municipal Securities Rulemaking Board (MSRB). In 1972–1973, the SEC created a new Market Regulation Division, the duties of which significantly overlapped those of Trading and Markets. Policies expressed in published regulations and no-action letters usually represent the consensus view of SEC staffers from both divisions.³⁷

In 1971, President Nixon appointed the well-connected New York lawyer William Casey as SEC chairman. The media reported that this role had been offered to other individuals who turned it down because they did not wish to lead an agency with low staff morale.³⁸ Casey's February 1973 departure from the SEC was marked by controversy because he had frustrated the Senate Judiciary Committee's access to SEC files related to connections between the International Telephone and Telegraph Corporation and the 1972 Nixon election campaign. Casey's successor was G. Bradford Cook, the former SEC counsel general and head of the SEC's newly created Market Regulation Division. Cook resigned just seventy-four days after his Senate confirmation because he was publicly associated with the criminal activities of financier and Nixon associate Robert Vesco. Although Cook avoided prison, he was disbarred, effectively ending his career in government.³⁹

Cook's successor, Ray Garrett, served from August 1973 until October 1975. A lawyer and SEC staffer from 1954 to 1956, Garrett achieved a degree of political independence from the administration that was unusual for a SEC chair and was likely due to Nixon's focus on the Watergate scandal.⁴⁰ Garrett expanded the SEC's role to include the investigation and punishment of U.S. corporations that corrupted foreign and domestic politicians. Corrupt practices unearthed by SEC officials during his tenure resulted in the 1977 passage of the Foreign Corrupt Practices Act. Garrett also expanded the SEC's workforce. His successor as chairman, Ford appointee Roderick Hills, in contrast, believed in deregulation and sought to limit the rapid growth of the SEC's workforce and responsibilities. Hills, it should be stressed, did not seek to modify the uniform net capital rule developed under Garrett or to interfere with the implementation of the NRSRO designation.⁴¹

Public discussion of the proposed uniform net capital rule began in early 1973, under Casey. The SEC official mentioned most frequently in the archived correspondence related to the proposed regulatory adjustment is Lee A. Pickard, the associate director of the Market Regulation Division. During the rule-making process, Pickard attended the conference of the Securities Industry Association and gave an interview about the proposed changes to Wall Street Journal reporters, who described him as the director of the Market Regulation Division.⁴² Pickard had recently been promoted to this position within the SEC. A specialist in securities law who left the New York law firm of Nixon Mudge Rose Guthrie Alexander & Mitchell, Pickard started work at the SEC shortly after Casey's appointment as chairman. (In the name of the firm, Nixon indeed referred to the president and Mitchell to his Attorney General John N. Mitchell.) Pickard later recalled that Casey, whose interests focused on geopolitics and intelligence, did not want “to get into the nitty gritty” of SEC regulations, preferring to leave details to Pickard, his “special assistant.”⁴³ Pickard's status as Casey's trusted protégé helps to explain the important role he played in the making of the uniform net capital rule in 1972 despite his relatively junior position in the SEC hierarchy at that time.⁴⁴

The SEC's Decision-Making Process

Although Section 8(b) of the Securities Exchange Act of 1934 authorized the SEC to limit the ratio between broker-dealer indebtedness and its “net capital,” the SEC first imposed a net capital rule on broker-dealers only in 1942.⁴⁵ This rule was analogous to those that the FDIC had recently imposed on banks in an attempt to ensure that their assets would be sufficient to cover their obligations in a crisis. The SEC rule stated that broker-dealers’ indebtedness could not exceed a ratio of 20:1 and specified that some classes of assets should be counted at less than their face value, a risk-weighting practice known as a “haircut.” During the first thirty years of Rule 15c3-1, the SEC independently determined haircuts and thus did not compel any broker-dealer to consider the ratings published by the bond-raters.⁴⁶ However, it exempted from its net capital rule members of the American, Boston, Midwest, New York, Pacific Coast, and PBW (Philadelphia-Baltimore-Washington) stock exchanges, which is to say almost all of the large broker-dealers, on the grounds that these exchanges imposed net capital rules on members that were more stringent than its own.⁴⁷

In 1939, the most important of the exempted exchanges, the NYSE, implemented a forerunner of its better-known Rule 325 of 1957, which mandated that its member broker-dealers determine haircuts using bond ratings assigned by “any of the nationally known statistical services.”⁴⁸ In 1971, the NYSE modified its rule book by specifying that in following Rule 325 exchange members should use the ratings of “Standard & Poors,” “the National Credit Office” (i.e., Dun & Bradstreet), or “any nationally known rating agency recognized by the Exchange.”⁴⁹ It appears, however, that the NYSE and other exchanges did not enforce these rules on members during the so-called back-office crisis of 1969–1970. The trust fund that the NYSE maintained to reimburse the customers of bankrupt broker-dealers came to the brink of failure in 1970. In the late 1960s, over one hundred broker-dealers, especially smaller ones, ceased to exist as independent entities because they could not afford the investments in people and technology needed to accommodate surging trading volumes. That crisis induced Congress to pass the Securities Investor Protection Act of 1970.⁵⁰

In August 1971, the SEC announced it was considering changes to Rule 15c3-1 that would impact the types of securities that broker-dealers would be required to hold, explaining that individual investors needed to be better protected in the event of broker-dealer insolvency. The SEC mentioned haircuts in this announcement only briefly and did not carefully distinguish between haircuts imposed as a result of credit ratings and those resulting from changing market prices.⁵¹ Speaking in November 1971, Kevin Duffy of the SEC's New York office explained that stronger protections for retail investors were needed in light of the “track record of this industry over the last 2½ years,” when shaky brokerage houses had imperiled customer funds.⁵² Congressman John Moss of California concurred, declaring that the Securities Investor Protection Corporation inadequately protected individual investors and that broker-dealers needed to be required to hold more capital.⁵³

At the end of 1972, the SEC released a detailed draft of its proposed net capital rule change, with Chairman Casey explaining that in making the rule the agency had drawn “heavily from its own experiences during the 1968–70 period” when at least fifty broker-dealers had failed, imperiling the savings of investors and damaging the commission's reputation as an effective guardian of retail investors.⁵⁴ Casey elaborated that the SEC wished to modify the net capital rule so that it applied to “all brokers and dealers whether or not they are a member of registered national securities exchanges.”⁵⁵ No references to bond rating agencies appeared in either the proposed net capital rule published by the SEC in late 1972 or in the associated press release from Casey's office.

Only during the ninety-day public consultation period that began with Casey's press release did references to bond rating agencies begin to appear in SEC documents.⁵⁶ Firms closely connected to the NYSE first proposed mandatory use of bond ratings to the SEC, effectively urging the SEC to impose NYSE Rule 325 on all broker-dealers in the United States. News of the proposed rule change agitated brokerage house owners across the country. The cost of complying with the SEC's proposed changes, which were designed to protect the roughly $2 billion in customer funds that broker-dealers typically held, was projected to be at least $223 million—too much for some firms in the “financially pinched securities industry” to bear without diluting their equity.⁵⁷

During this first consultation period, the commission received 151 letters regarding the proposed changes to Rule 15c3-1.⁵⁸ While some of the letters were addressed to Cook, the director of the SEC's Market Regulation Division, a majority (132) went to Pickard, Cook's nominal subordinate and Casey's right-hand man. Several smaller firms expressed fear that the new version of Rule 15c3-1 would advantage larger firms better able to bear the costs of complying with the new rules, which the correspondents regarded as more burdensome than existing SEC regulations. These correspondents strongly objected to the proposal on the grounds that it effectively involved extending NYSE Rule 375 to small firms unable to suffer the high compliance costs. One upstate New York correspondent argued that the proposed net capital rule “appears to be unfair to small broker-dealers” as it would limit such firms to making markets in just “15 higher priced securities.” That correspondent also predicted that while the rule revisions were ostensibly designed to prevent broker-dealers from becoming insolvent, the proposal could “very well lead to the insolvency of firms such as ours through the attrition of profits” and higher overhead costs.⁵⁹ Another broker-dealer also predicted the proposed change “will virtually force all small broker-dealers out of business.”⁶⁰ Correspondents in Florida and Oregon pleaded with the SEC to consider the impact of the proposed rules on “small brokerage houses” and the “small [issuing] companies that need such brokerage houses to make a market for their securities.”⁶¹

Thirteen letters received by the SEC at this time explicitly referred to the bond ratings agencies. Of those, nine explicitly recommended that the SEC's new net capital rule mandate the use of agency ratings. In its letter to Pickard, the National Association of Securities Dealers (NASD) recommended that the SEC require firms to follow the NYSE by requiring broker-dealers to use ratings produced by “any nationally known statistical service which is recognized by the SEC.”⁶² The letter neither named the firms it considered to be nationally known statistical services nor suggested any criteria for deciding which bond rating agencies the SEC should designate as such.

Letters to the SEC that referred to agency ratings were written when press coverage of, and congressional attention to, the agencies and their leaders had increased. Prior to the late 1960s, the personnel and the procedures of the ratings agencies attracted little attention from Congress or the leading newspapers.⁶³ The ratings agencies had, as it were, flown under the radar of public opinion. After about 1967, references to the ratings agencies became more frequent in the press and Congress, and their leaders, S&P president Brenton W. Harries and Jackson Phillips of Moody's, rose in profile.⁶⁴ Their profile further rose in the aftermath of the agencies’ shift to the issuer-pay model, which was implemented in steps between 1968 and 1970, and the controversial downgrading of the debt of such cities as Trenton and New York City.⁶⁵

Members of the Joint Economic Committee of Congress held hearings into the rating agencies in December 1967 and July 1968, which allowed municipal officials to criticize the ratings as “subjective,” produced by firms with insufficient manpower and computer resources, and unfair to some municipal issuers.⁶⁶ During the hearings, rating agency representatives refuted accusations that each rating represented just “ninety minutes” of analyst time by describing the elaborate procedures by which teams of academically qualified analysts and their computers produced ratings.⁶⁷ Notwithstanding such disclosures, critics continued to denounce agency ratings as subjective. In 1973, for example, Congressman John Murphy, who represented a New York City district, demanded regulation of the bond ratings agencies that had, in his view, unfairly raised the borrowing costs of his city.⁶⁸

Ratings agencies also faced criticism from Wall Street “investment analysts,” the staff of the FDIC, and Federal Reserve Board member George W. Mitchell, all of whom suspected that the alphanumeric bond ratings were poor predictors of issuer default.⁶⁹ The newly controversial nature of the bond rating agencies helps to explain why supporters used somewhat tentative language in their submissions to the SEC regarding the use of agency ratings in haircut determinations. The law firm of Davis Polk & Wardwell (DPW), for example, cautiously endorsed the use of ratings in determining net capital requirements. One of the most powerful Wall Street law firms, DPW in its letter to Pickard discussed the advisability of including a reference to bond ratings but was less forceful than the NASD. After noting the disparity between the proposed SEC haircuts and the existing NYSE schedule, DPW urged the SEC to copy the NYSE rules. Doing so would mean requiring all broker-dealers to rely on the opinions expressed by ratings agencies: “We suggest that the Commission consider lower capital charges for commercial paper of issuers whose registered debt securities have satisfactory ratings from qualified rating services.”⁷⁰ Like the NASD, DPW did not explicitly name the ratings services that its clients considered qualified to rate securities.

The idea that the SEC's net capital rule should also refer to bond ratings was also discussed in a memorandum submitted to Pickard by Arnold H. Tracey, a partner at Mudge Rose Guthrie Alexander & Mitchell, the New York law firm that had previously employed Pickard, President Nixon, and Attorney General Mitchell. The memorandum the Mudge lawyers had prepared “on behalf of several of the major commercial paper dealers” in New York recommended that the SEC's net capital rule be modified so that it more closely matched the net capital rule the NYSE imposed on its members. The memorandum concluded ambiguously that “while the dealers as a group believe that the continuation of ratings as a part of the net capital test is desirable, we recognize that regulatory philosophy [emphasis added] may dictate against their use,” so, “accordingly, we are not recommending their inclusion in any new proposal.”⁷¹ The letter did not specify the meaning of “regulatory philosophy,” a term not delineated in contemporaneous documents in the SEC archive or press. While we are thus unable to determine with certainty what this term meant to the letter's author, he may have been alluding to the fact the use of bond ratings was controversial among SEC staffers.

After the ninety-day public consultation period closed, SEC staff in the division of Market Regulation and the nine regional offices had an internal discussion of the details of Rule 15c3-1. Some SEC officials disagreed with the proposal made by DPW and the NASD that the future regulation should mandate the use of bond ratings. Marc Berman of the Market Regulation Division, for instance, told George M. Callahan of the Atlanta Regional Office that “the opinion of the Division” was that for “the treatment for net capital purposes of prime commercial paper and negotiable certificates of deposit,” the current market value of securities, not bond ratings, was preferred. Berman himself regarded “marking to market” as superior to reliance on Big Three ratings.⁷²

Other SEC staffers strongly concurred with Berman's view that ratings were too flawed to be explicitly endorsed by the SEC. The 1972 SEC investigation into the collapse of Penn Central revealed years of mismanagement by the railway's managers and, crucially, that observers of the firm in the relevant division of Moody's Investors Services were, by 1969 at the latest, aware that the railroad might soon declare bankruptcy. The National Credit Office (NCO), the division of Moody's Investor Services that rated commercial paper, had nevertheless rated Penn Central Transportation Co. securities as investment grade or “prime” until just three weeks before the bankruptcy. According to the SEC's report, the rating agency's “concealment of Penn Central's condition . . . contributed to the misleading of investors” who continued to purchase Penn Central obligations well after February 1970, when managers at Goldman Sachs and NCO realized that Penn Central faced imminent bankruptcy. The report named and criticized ratings agency employees like Rudolph G. Merker and Allen Rogers for their role in the debacle, which included informing Jack Vogel of Goldman Sachs, but not the investing public, of Penn Central's problems. After Rogers informed Vogel of the railroad's problems, Goldman Sachs accelerated its sale of Penn Central securities to unsuspecting retail investors. The 1972 report demonstrates that individuals within the SEC understood the conflicts of interest associated with the Big Three's issuer-pays model and that Big Three bond ratings were highly fallible guides to the future.⁷³ When he was asked in September 1973 whether the SEC's new net capital rule would, in fact, mandate the use of the bond ratings produced by agencies such as Moody's and S&P, Pickard reported that SEC staffers were “still not certain about that” as doing so “would take us away from our traditional concept of never commenting on a security.”⁷⁴

Less than two months later, the SEC announced that the uniform net capital rule would indeed mandate the use of such ratings by all broker-dealers. It did so on November 29, 1973, when it published a substantially modified proposal that mandated that haircuts for some securities would depend in part on ratings provided by “at least one of the nationally recognized statistical rating organizations.”⁷⁵ During the subsequent consultation period, the SEC received 143 letters, 8 of which referred to the use of NRSRO ratings. The letter from prominent New York law firm Sullivan and Cromwell noted that the proposed haircut schedule would “discourage” the trading of foreign government securities in the United States because “relatively few” of those securities were rated by the bond rating houses.⁷⁶ The SEC staff likely paid considerable attention to this letter, given that Sullivan and Cromwell had played an important role in international finance since the establishment of its first European office in 1911. For instance, the firm had helped issuers in the Weimar Republic to market securities in the United States, a project designed to help bolster Germany's new democratic political system. More recently, two of the firm's partners had served in the Eisenhower administration, one as secretary of state and the other as head of the Central Intelligence Agency.⁷⁷ The firm's interests had long been intertwined with U.S. foreign policy. As it helped foreign firms to issue securities on Wall Street, Sullivan and Cromwell had an obvious reason to oppose a rule that would have discouraged trading in such securities. For reasons that remain unclear, SEC staffers did not modify the proposed rule to accommodate the concerns of Sullivan and Cromwell.

In its letter to the SEC, a New York broker-dealer argued that highly rated securities should be exempted from a proposed concentration rule.⁷⁸ The representative of a Chicago broker-dealer firm argued that the proposed rule's reference to NRSROs should “be further clarified” by actually naming the firms within this category. He suggested Standard and Poor's, Moody's, and Fitch.⁷⁹ In a very lengthy memorandum on the likely impact of the new net capital rule on the investment industry, Oppenheim, Appel, Dixon & Co. distinguished between the impact on broker-dealers that were members of the NYSE and those that were not. The firm's analysis suggested that implementation of the revised proposal would have a largely neutral effect on NYSE members, for while SEC Rule 15c3-1 differed in some respects from NYSE Rule 325, “generally the SEC haircuts are no more severe.” The impact on broker-dealers who were not NYSE members would, according to the memorandum, be more considerable, as their maximum capital ratio would be reduced from 20:1 to 15:1 and “for the first time, non-members would be required to haircut exempt securities in computing net capital.”⁸⁰

On June 26, 1975, the SEC published the long-awaited final text of the new net capital rule and announced that it would go into effect in September. The final text of the rule declared that in determining the haircut to be applied to a security held by the broker-dealer, the ratings generated by at least two NRSROs should be consulted. The earlier draft of the rule had mandated the use of just one NRSRO rating. The SEC did not explain this adjustment in the rule, but it may have been designed to check ratings inflation. In publishing the new rule, the SEC specified neither the identities of the bond ratings agencies it deemed to be NRSROs, as had been proposed by the correspondent in Chicago, let alone its criteria for determining which agencies should have NRSRO status. The SEC did not name the agencies it considered to be NRSROs until the publication of a no-action letter to Lehman Brothers on March 18, 1976, explicitly mentioning the Big Three.⁸¹

The volume of letters that the SEC received during both consultation periods suggests that reform of the net capital rule was important to broker-dealers and other SEC-regulated firms. The modification of the net capital rule that required broker-dealers to use the ratings produced by NRSROs did not, however, attract the attention of members of Congress or editorial writers at publications like the New York Times, the Washington Post, and the Wall Street Journal. Although New York City's fiscal crisis in the fall of 1975 prompted observations in the press about the unreliability of bond ratings, it does not appear to have raised concerns about the SEC's recent decision to mandate the use of the ratings produced by the NRSROs. In his October 1975 testimony about the New York City default before a congressional committee, economist Robert D. Reischauer remarked that “some responsibility must rest with the bond rating agencies who chose to upgrade the ratings” of a city that was “running a large current account deficit and accumulating a huge amount of short-term debt.” However, Reischauer said nothing about the regulation the SEC had introduced in June 1975 that effectively required all U.S. broker-dealers to consult NRSRO ratings.⁸²

In June 1976, Jackson Phillips of Moody's and S&P president Brenton W. Harries were called before a congressional committee to account for the fact that their firms had given New York City bonds a high rating until shortly before the city's financial crisis.⁸³ Phillips and Harries had been summoned by members of Congress who advocated requiring the SEC to establish an appeal procedure whereby municipalities that had been downgraded by bond rating agencies could request that the SEC adjust the rating.⁸⁴ Responding to this proposal, SEC commissioner Phillip Loomis noted that the agency did not wish to be involved in judging the opinions contained in NRSRO ratings, as doing so would “inject the agency” into the vexed question of which municipal governments were most at risk of default.⁸⁵

Other witnesses testified that the idea of creating a “Federal Rating Agency” to replace the for-profit bond rating agencies, which had circulated in Washington since the late 1960s, would involve the adoption of an unacceptable European-style “central type of bond system . . . [that] would be a complete repudiation of state and local sovereignty.”⁸⁶ These witnesses attacked the proposals to nationalize the task of rating securities as fundamentally un-American and quasi-socialist. Shortly after the hearings, the proposal to make the SEC an appellate body for municipal bond ratings was dropped. At no point during these hearings did any speaker refer to the fact the SEC had recently introduced a regulation that required firms to use the ratings produced by the Big Three.

In a 1977 internal report, SEC officials observed that until shortly before the collapse of New York City's finances, the NRSROs had maintained high ratings on its debt and that their ratings had exerted an “enormous impact” on investment decisions. The report declared that both Moody's and S&P had “failed, in a number of respects, to make either diligent inquiry into data which called for further investigation, or to adjust their ratings of the City's securities based on known data in a manner consistent with standards upon which prior ratings had been based.” The report did not, however, propose reviewing either the NRSRO status of those bond rating agencies or the 1975 rule that mandated the use of their ratings.⁸⁷

In 1978, a mystified British financial journalist observed that Wall Street firms continued to accord considerable credibility to bond ratings despite their dubious predictive value. He observed that Moody's and Standard and Poor's gave “Penn Central a high rating before it collapsed” and similarly rated New York City paper as investment grade until shortly before the “Big Apple ran into trouble.” “Who rates the raters?” he asked. The journalist, who observed that decision-makers in London investment firms accorded little credence to bond ratings, did not appear to consider the possibility that American investment firms were continuing to pay attention to these ratings because federal regulation now required them to do so.⁸⁸

One striking feature of the process by which the NRSRO designation became part of Rule 15c3-1 is that neither the advocates nor the opponents of mandating the use of NRSRO-generated bond ratings to determine haircuts appear to have thought about the ratings agencies through a competition policy lens. The 1960s and 1970s are today remembered as the “golden age of antitrust” enforcement and as an era when the Department of Justice (DoJ) frequently launched antitrust action against firms, including quite small firms that were accused of being excessively dominant in narrowly defined markets. This era is frequently contrasted with the period after 1980, when the growing influence of Chicago School ideas inhibited antitrust action.⁸⁹ In the 1960s and 1970s, only two firms dominated bond rating, but DoJ lawyers did not even propose antitrust action against them.⁹⁰ Moreover, neither the staff of the SEC nor officials in any other government department expressed concern that the design of the net capital rule regulation might discourage healthy competition. After 2003, policymakers and academics did begin to think about using competition policy to address the evident problems with the NRSRO ratings, but this idea was not expressed in the period covered by the present study.⁹¹ The striking failure of SEC policymakers to think about the implications of the 1975 changes to SEC Rule 15c3-1 for the level of competition in the ratings industry is perhaps best explained by concluding that the commission did not, at least in this period, see promotion of competition as part of its mandate. The SEC was, for example, reluctant to move against fixed stock commissions, a clearly anticompetitive practice eliminated only in May 1975.⁹²

Conclusion

This study of the making of the uniform net capital rule and the origins of the NRSRO designation reveals a pattern broadly congruent with the theory of bureaucratic imperialism. Political scientists have argued that a regulatory agency that issues vague rules maintains greater regulatory discretion than one that constrains itself by promulgating unambiguous rules. The 1975 decision of the SEC to require firms to use NRSRO ratings increased its regulatory discretion. During the period in which it was formulating the uniform net capital rule, from 1972 to early 1975, the SEC was operating in a condition of political slack. The spectacular financial problems of New York City in 1975 heightened public interest in the issue of bond rating agencies and their conflicts of interest, thereby producing, in our view, a change in the behavior of the regulators. As predicted by the theory developed by Zachary Gubler, the SEC in 1976 refrained from expanding its power over ratings by opposing a proposal in Congress that would have made it a ratings appeals tribunal.

The behavior of the SEC during this episode can also be interpreted as the result of individuals acting in a path-dependent and boundedly rational manner.⁹³ In the early 1970s, the SEC faced external pressure to develop a system, preferably a labor-efficient one, for established bond rating haircuts and it simply adopted a system, which involved the consultation of alphanumeric bond ratings, that other regulatory agencies had already adopted without incurring too many costs. This action solved an immediate problem but created future ones that only became manifest after the key decision-makers had left the SEC.

These findings help to identify directions for future historical research. First, smaller broker-dealer firms complained during the consultation period that the SEC's proposed uniform net capital rule would drive them out of business by raising their overhead costs. They argued that while larger firms could easily cope with the costs associated with the proposed regulation, smaller firms, which were already under intense pressure, could not. After 1975, the securities industry indeed experienced considerable consolidation, with many smaller broker-dealers either disappearing or being absorbed into larger entities.⁹⁴ The acceleration of this trend in recent years has been attributed to the steadily growing burden of SEC regulations, which disproportionately affect the overhead costs of smaller broker-dealers.⁹⁵ Of course, this chronology does not prove that the 1975 uniform net capital rule contributed to the consolidation of the industry, as consolidation might have occurred as a result of nonregulatory actors such as changes in technology. Only rigorous econometric research would permit one to determine whether the 1975 uniform net capital rule, or any other SEC regulation that raised the costs of smaller broker-dealers, contributed to consolidation.

The other avenue of research is the need for more extensive study of the role of the SEC in shaping the laws that help to make America's corporate governance system different from those in which managers are regarded by many academics as more attentive to the needs of non-shareholder stakeholders.⁹⁶ Qualitative historical research methods are particularly suitable for conducting this program of research, which would blend legal, political, and business history. Joel Seligman's history of the SEC appeared in 1982 and was informed by the political-economic debates of that time.⁹⁷ We need historical research on the SEC that speaks to today's issues, such as the SEC's increasingly controversial quarterly reporting requirement. Since 1970, the SEC has required firms to issue quarterly reports to investors. Critics argue that this rule promotes a short-term mentality in American boardrooms that discourages U.S. companies from making long-term investments of the type associated with German and East Asian firms. The SEC's quarterly reporting requirement has been critiqued by diverse scholars and even, on one occasion, by an incumbent U.S. president.⁹⁸ Scholars know little, however, about the SEC's motives for introducing this important rule.⁹⁹

Similarly, business historians currently know little about the extent to which the SEC contributed to the renaissance of so-called “shareholder value ideology” in the 1980s.¹⁰⁰ In recent years, the doctrine of shareholder primacy has been assailed by a range of academic and non-academic authors who argue that when a nation's managers believe that their sole duty is to maximize shareholder value, society suffers.¹⁰¹ In August 2019, the Business Roundtable, a group of leading U.S. executives, explicitly repudiated shareholder primacy and endorsed a stakeholder-centric vision of corporate purpose.¹⁰² The legal and management academics who criticize shareholder primacy typically argue that from the New Deal to the early 1980s, American companies were not managed in accordance with the principle of shareholder primacy and that the business executives of this period saw “themselves as stewards or trustees charged with guiding a vital social and economic institution in the interests of a wide range of beneficiaries,” not just shareholders.¹⁰³ As Alex Edmans has noted, authors who attack shareholder primacy almost invariably denounce Milton Friedman's September 1970 New York Times piece that declared that the sole social responsibility of business is to increase profits. They also tend to suggest that the publications of Friedman and like-minded economists contributed to the shift in the 1980s to shareholder primacy.¹⁰⁴ Although we might debate whether the writings of Friedman were as important a causal factor as is sometimes asserted, it is clear that there was a renewed emphasis on shareholder value maximization in the 1980s associated with a more competitive market for corporate control and the rise of the leveraged buyout fund.¹⁰⁵ Although the role of the SEC in shaping the relationship between managers and shareholders is widely acknowledged, scholars know little about whether and how SEC rule changes in the 1970s and 1980s affected the return of shareholder primacy.¹⁰⁶ Moreover, we do not know what SEC officials in this era thought of the arguments of the economists and others who advocated a return to shareholder primacy. Clearly, a new wave of historical research on the SEC that uses archival sources and oral history to answer such questions could deepen our understanding of the origins of the distinctive features of the American national business system.