By Henri Servaes, Ramin Baghai, and Ane Tamayo
In the wake of the financial crisis, rating agencies came under attack. They were accused of having been too generous when rating mortgage backed securities and other structured finance products. It became common wisdom that they had not been critical enough in making their judgements, but the reality is more subtle.
Our research shows that in recent decades, agencies have actually become more conservative when rating corporate debt. Corporate bonds are a different asset class from the structured debt securities that were at the centre of the financial crisis, and they constitute the “bread and butter” of rating agencies’ businesses. This conservatism has had serious implications for businesses that now find it harder to borrow money than in previous decades.
So why does the post-crisis accepted wisdom conflict so sharply with our findings? Is there confusion about the type of debt being discussed? We believe so. The ratings of asset-backed securities such as the collateralised mortgage obligations that sparked the credit crunch appeared inflated: they may not have been as sound as the rating agencies said. But the ratings awarded to corporate bonds – the subject of our research – have moved in the opposite direction. From 1985 to 2009, average ratings declined by three notches. For example, a firm with a AAA rating in 1985 would only qualify for a AA- rating by 2009, holding all its characteristics constant. This has consequences, both for the companies concerned and for the wider economy.
Credit rating inflation – where agencies increasingly hand out good ratings to debt securities that are actually quite risky – has been the conventional wisdom since the 2008/09 crisis. It is an understandable notion and dislodging it will not be easy. The most serious claim is that because companies paid for their own credit rating, there was a conflict of interest: credit rating agencies relaxed the standards to keep their customers happy, helping create the conditions for the financial crisis. This accusation has spurred regulators and legislators on both sides of the Atlantic to scrutinise credit rating agencies more closely.
The truth about credit rating agencies
But the idea that rating agencies have, over decades, become less critical in assessing the quality of companies’ debts is simply at odds with reality. The truth is that the agencies have actually become more conservative when rating corporate bonds during the last 30 years rather than more lax. Why? This increased conservatism cannot be explained by corporate credit becoming riskier. It has not: bond default rates have actually declined across the board. Increased conservatism is seen among both investment-grade and non-investment-grade bond issuers.
So how did we identify this apparent trend towards greater conservatism among credit rating agencies? Put simply, we modelled ratings awarded by Standard and Poor’s, one of the main rating agencies, from 1985 to 1996. We then used that model to predict the ratings that would have been awarded to corporate debt issued from 1997 to 2009 if the standards from the earlier period had been applied. From there, we looked at the ratings that had actually been awarded between 1997 and 2009 and compared the two. We found that the average corporate bond from 1997¬ to 2009 received a rating several notches lower than it would have been if assessed using the 1985 to 1996 standards.
While the major credit rating agencies – Standard & Poor’s, Moody’s and Fitch – vary slightly in terms of their categories, credit ratings generally run from AAA for debt with the lowest credit risk right down to D, which refers to bonds whose issuers are in default. Rating categories are also fine-tuned to reflect small changes in credit risk: for example, a AA+ rating denotes lower credit risk than a AA rating, which in turn reflects lower risk than AA-. Ratings between AAA and BBB- are described as “investment grade”, while categories from BB+ to C signify varying degrees of “high yield” or “junk” status. These categorizations matter for investment mandates, financial regulation, and capital requirements of insurers and banks.
Each movement of one category up or down, such as from AAA to AA+, is described as one notch. The rating given to corporate debt by agencies is very important to the companies concerned. As a rule, the lower the rating, the higher the cost of borrowing for the issuing company. It is reasonable to infer that increasing conservatism among rating agencies in past decades would have very serious consequences for a borrower with a low investment-grade status.
Take a company whose bonds were rated BBB in 1985. Our research shows that if the bonds of the same company with the same fundamentals had been assessed by a ratings agency 20 years later, they would no longer have achieved the same rating. Instead, the bonds would have been classed as “junk”. The company itself was no less creditworthy in 2005 than it was during the heyday of Ronald Reagan and Margaret Thatcher, but it was assessed by the rater as being less safe.
Our research suggests that capital markets have not entirely overlooked the rating agencies’ increased stringency. Lenders have made some allowances for the agencies’ move to more demanding standards. But those allowances are not big enough: the shift in capital markets has not been sufficient to offset fully the agencies’ tougher approach.
How greater stringency affects businesses
What effect does this have on both the companies concerned and the broader economy? Perhaps the most immediate impact is that the businesses affected issue less debt than would otherwise be the case. Take a company whose debt rating is pushed down a notch as a result of the rating agencies taking a more stringent approach: on average, that company’s debt issuance as a share of its assets decreases by 8%. So leverage decreases among those firms most affected by the new conservatism of the rating agencies. These companies also tend to hold more cash and, critically for the economy as a whole, they experience slower growth.
There is better news in terms of investment, where the impact of greater stringency seems to be modest. But it is a different story when it comes to cash acquisitions of other companies: there is a substantially negative effect.
All of this leaves the unanswered question of why the credit rating agencies have adopted a more conservative approach. One possibility is that the rating agencies had been too lenient in the past and subsequently sought to put this right. In other words, there had been some rating inflation, but it was later corrected.
But our research suggests that greater conservatism among agencies is not simply to address their previously lenient approach. Nor is it a response to rising default rates. We find that companies whose ratings were affected more by increased rating agency conservatism were actually awarded lower spreads by the market. This would not have happened had the conservatism merely been a response to past leniency.
Identifying the motivation for this increased stringency is outside the scope of our research. But one possible explanation is that the agencies tightened their policies in the wake of high-profile corporate collapses and dotcom businesses going bust at the start of the Millennium. Even so, that does not explain why the increased stringency has taken place since the mid-eighties.
Since the financial crisis, it has become commonplace to question how credit ratings are awarded and to suggest that agencies’ methods and policies need to be more fully investigated. We agree that this is likely to have been the case with structured finance products, where conflicts of interest are arguably more severe than in other debt categories. However, a less discussed problem has been excessive conservatism in assigning corporate debt ratings, a conservatism that has real-world consequences for businesses.