Credit Rating Agencies: History and Role in Financial Crises

Credit rating agencies transformed from obscure bond research boutiques into gatekeepers of global capital markets — then catastrophically misjudged structured finance. Understanding their rise, the incentives that broke their judgment, and their continuing role in regulation reveals why ratings matter and why they fail.

From railroad bonds to market gatekeepers

When Moody’s published its first railroad bond ratings in 1909, the business was simple: a subscriber paid for a handbook, looked up a bond, and saw a letter grade. The credit rating agency earned a subscription fee, not from the bond issuer. Impartiality was baked in because the buyer of the rating — the investor — paid the bill.

For decades, this remained niche work. Banks, insurers, and large portfolio managers subscribed. The three agencies that would eventually dominate — Moody’s, Standard & Poor’s, and Fitch — operated as semi-monopolies, but each had limited reach and faced honest competition from credit analysis done in-house by large institutions.

Everything pivoted in the 1970s. The SEC, worried about fraud and fragmentation, designated certain agencies as Nationally Recognized Statistical Rating Organizations (NRSROs). This meant their ratings could be cited in regulatory filings. Banks and insurance companies suddenly needed an NRSRO rating to satisfy regulators. The three major firms became necessary — not optional.

But the regulatory blessing came with a hidden cost: it shifted power toward the issuers. If a bank or corporation needed a rating to raise capital, and the rating had to come from one of three gatekeepers, those firms could now demand the rating and negotiate terms. By the 1990s, most agencies had switched to an issuer-pays model, where the company raising capital paid the rating agency to analyze the bond.

This inversion — from investor-pays to issuer-pays — was the structural flaw that would later feed the 2008 crisis.

The structured-finance trap

In the 1990s and early 2000s, securitization exploded. Banks would originate mortgages, bundle them, slice them into tranches, and sell them as mortgage-backed securities (MBS). Each security needed a rating.

Here’s where the incentive structure broke down. A rating agency that refused to rate a mortgage bond would lose the fee; a competitor would do it instead. If Agency A rated a tranche BBB and Agency B rated it AAA, the issuer would use B’s rating (and pay B). This was “rating shopping” — issuers found the most generous rater.

Worse, structured finance was complex. The models required to rate a mortgage bond were opaque. Agencies employed quants who built models, but the data feeding those models — mortgage defaults, housing price paths — was drawn from a period of extraordinary conditions. Home prices had risen steadily for decades. Mortgage lending standards had eroded. The agencies modeled under the assumption that the boom would continue or at least stabilize, not that prices could collapse nationally.

The agencies also faced a technical problem: they had no historical data on mortgage pools of this size and diversity. They borrowed from equity-index models and simplified assumptions. A AAA-rated mortgage bond, the reasoning went, should default at the same rate as a AAA-rated corporate bond. But the mechanism was utterly different. A corporate bond defaults when a mature firm’s earnings crater; a mortgage bond defaults when borrowers lose jobs or home values collapse. The correlation structures were misunderstood.

By 2006, the three agencies had rated mountains of mortgage securities AAA — the highest rating, implying virtually no default risk. Many of those bonds failed within two years.

Structural reform and lingering dependence

The collapse of 2008 destroyed the agencies’ credibility. Moody’s, S&P, and Fitch faced lawsuits, regulatory fines, and public fury. Congress passed the Dodd-Frank Act in 2010, which created an Office of Credit Ratings within the SEC and required agencies to file internal control reports. The law also attempted to reduce regulatory reliance on ratings — the idea being that if regulations didn’t mandate the use of a rating, issuers couldn’t shop for the best one.

This attempt largely failed. The agencies remain integral to banking regulation. Tier 1 capital rules, leverage-ratio haircuts, and bond-eligibility criteria for collateral still reference credit ratings. A downgrade of a sovereign or bank can trigger rapid capital calls. Regulators said they wanted to reduce ratings dependence but couldn’t design systems without it.

The issuer-pays problem was never fully solved either. Agencies were asked to add safeguards — rotating analysts, transparency about models, conflict-of-interest disclosures. But the economic incentive remains. An issuer dissatisfied with an agency’s analytical approach will gradually move business to a competitor.

What changed is reputational cost. A major ratings failure now brings fines, subpoenas, and investor lawsuits. Agencies have tightened models for structured finance. But they still face the same trade-off: be conservative and lose market share, or stay competitive and risk rating another bubble.

Why ratings still matter despite distrust

Despite their failures, credit ratings still shape capital flows. A company’s credit spread — how much above the risk-free rate investors demand — is strongly tied to its rating. A downgrade from A to BBB can add 100 basis points to borrowing costs overnight. High-yield bonds (often called junk bonds) by definition carry a sub-investment-grade rating, and many investors are legally barred from holding them.

Insurance companies, pension funds, and banks are required to hold a certain amount of investment-grade debt. A rating downgrade by a major agency can force these institutions to sell, creating fire-sales and price pressure.

Ratings also embed information that markets may not immediately price. A downgrade often precedes actual default risk by months. The agencies, for all their failures, employ large teams of credit analysts who spend time understanding borrower financials, industry dynamics, and covenant structures. A solo retail investor or trading desk may not replicate that analysis.

The ongoing tension

Twenty years after the crisis, the tension between ratings’ utility and their fallibility remains unresolved. Regulators want to depend less on them but can’t; issuers want low ratings but pressure agencies to provide them; investors need the information but must assume no single rating is definitive.

The major agencies have diversified beyond ratings. Moody’s owns MCO (a data and analytics giant). S&P is part of McGraw-Hill. Fitch is owned by Hearst. This has reduced their dependence on ratings fees but has not eliminated the conflict.

For credit analysis, ratings are a starting point, not a verdict. A AAA rating tells you an agency sees minimal default risk, but you should ask: What did they assume about future conditions? What assets back the bond? Who loses money first if the borrower struggles? Those questions demand independent work.

See also