End the Credit Rating Addiction

By John Kiff

One of the earliest take aways from the global financial crisis was the importance of access to information for effectively functioning financial markets. And, in that regard, credit ratings can serve an incredibly useful role in global and domestic financial markets—in theory.

In practice, credit ratings have inadvertently contributed to financial instability—in financial markets during the recent global crisis and more recently with regard to sovereign debt. To be fair, the problem does not lie entirely with the ratings themselves, but with overreliance on ratings by both borrowers and creditors.

In one of the background papers for the Fall 2010 Global Financial Stability Report that I prepared with IMF colleagues, we recommend that regulators should reduce their reliance on credit ratings. Markets need to end their addiction to credit ratings.

Credit ratings should be seen as one of several tools to measure credit risk, and not as the sole and dominant one. Instead, credit ratings, which measure the relative risk that an entity such as a government or a company will fail to meet its financial commitments, have become hardwired into various rules, regulations and triggers. Central banks often use ratings in their collateral acceptability rules. The Basel II standardized approach to determining bank capital requirements relies heavily on ratings. And, many institutional investors—pension funds, insurance companies, retirement funds, and the like—have rules that trigger the sale of securities when they are downgraded below certain levels.

Yet, when investors, regulators and borrowers rely on credit ratings too mechanically, changes in ratings, particularly abrupt downgrades, can lead to deleterious selloffs of securities. Not only does this create the potential for broader spillovers, but the resulting declines in the prices of securities trigger further sell-offs. In other words, those notorious domino effects!

Actions by ratings agencies to “smooth out” or make changes in ratings less abrupt—for example, through a warnings ahead of a possible downgrade—may be intended to minimize disruptions. However, they we find that much of the market reaction occurs when these warnings are released rather than when the actual rating changes.

And, as events of the past year or so have highlighted, sovereign ratings could have taken better account of debt composition and contingent liabilities. However, in some cases—Greece for example—the rating agencies did not have access to all the information they needed. In that regard, the IMF encourages countries to prepare and make publicly available a fiscal risk statement.

Still, the real solution lies in reducing the reliance on credit ratings as much as possible. This should start with removing the mechanistic use of ratings in rules and regulations, which some countries are already beginning to do. Investors must be weaned off credit ratings too. Policymakers should persuade the larger ones, at least, to perform their own risk assessments as part of deciding what to buy or sell.

Realistically, of course, not all investors have the same ‘in house’ capacity for risk assessment. Smaller and less sophisticated institutions will have to continue to rely heavily on third-party ratings. So, the process of reducing reliance on ratings should differentiate according to the size and sophistication of institutions, and the instruments being rated. Also, agencies—the main ones being Standard & Poor’s, Moody’s, and Fitch—whose ratings continue to play key regulatory roles (as in the Basel II standardized approach) should be subjected to increased oversight. (Both of these approaches were included in the recently signed U.S. financial sector reform legislation.)

Beyond this, our paper recommends that policymakers should also continue to push rating agencies to improve their procedures, including those related to transparency and governance. This will provide more assurance to those that use ratings that they are fairly constructed. Credit rating agencies aggregate information about the credit quality of various types of borrowers and their financial obligations. The ratings they issue allow many of those borrowers to access global and domestic markets they would not otherwise have, enabling them to attract investment funds. As a result, ratings add liquidity to markets that would otherwise be highly illiquid.

Wringing out the volatility, without drying up the liquidity, is the goal of any effort to reform the credit ratings agencies that issue them.

2017-04-15T14:33:42+00:00September 30th, 2010|


  1. Per Kurowski September 30, 2010 at 10:30 am - Reply

    In May 2003, as an Executive Director of the World Bank, I told those many present at a risk management workshop for regulators the following with respect to the role of the Credit Rating Agencies. “I simply cannot understand how a world that preaches the value of the invisible hand of millions of market agents can then go out and delegate so much regulatory power to a limited number of human and very fallible credit-rating agencies. This sure must be setting us up for the mother of all systemic errors.” And this I repeated over and over again, even in the press, even in formal statements at the Board.

    Now as reported the IMF is finally admitting “Policy makers should work towards the elimination of rules and regulation that hardwire buy or sell decisions to ratings”

    That is good, better late than never. But the real question that needs an answer is why on earth it had to take a financial crisis of monstrous proportions to reach a conclusion that should have been apparent to any regulator from the very beginning.

    I saw it happen in front of my eyes and I know why it happened. As I wrote in a letter published in the Financial Times in November 2004, it was the result of the whole debate about bank regulations being sequestered by the members of a small mutual admiration club.

    Therefore if there is now something even more important than rectifying the faulty financial regulations, that is to break up the Basel Committee and make absolutely sure it represents a much more diversified group of thinkers. That would have at least guaranteed that the basic question of what the purpose of the banks should be would have been put in the forefront before regulating them. Current regulations do not contain one word about that.

    Besides me there were plenty of experts who raised the question of whether the credit rating agencies should have such a prominent role and made many other valid criticisms. These persons who saw it should participate in designing and putting in place the needed reforms. It is simply unacceptable that the reforms that carry with them such huge global implications are implemented exclusively by Monday morning quarterbacks.

  2. Michael Pomerleano September 30, 2010 at 12:06 pm - Reply

    Re credit rating agencies, I doubt that urging reforms etc will help with a structural problem, as I pointed out in my article in the FT Economists Forum : “Clearly this market structure is prone to distortions. We are down to four big national accounting firms and three rating agencies. The rating agencies and accounting firms know that they have the regulators and financial industry over a barrel, because the regulators will not dare to provoke their demise and the end of transparency. This situation is not conducive to reforms.”

    See The Lehman bankruptcy examiner report: And then there were none Tuesday, April 6th, 2010 FT By Michael Pomerleano

    So in conclusion, I am skeptical that any set of measures will be effective when the rubber hits the road.

  3. Marco Moreno September 30, 2010 at 1:31 pm - Reply

    Rating agencies deliver a highly subjective opinion and are technically lax. They also have a strong dependence on the bench. They lack transparency and have no objective parameters

  4. Per Kurowski October 1, 2010 at 10:26 am - Reply

    Marco Moreno, the problem with the current regulatory paradigm goes much deeper than that of the credit rating agencies not doing their job right. The real fundamental problem is that the bank regulators took upon themselves to decide that the only thing of importance for the banks was to avoid the risk of defaults… and human development is so much more than that.

    In the “Explanatory Note on the Basel II IRB Risk Weight Functions” it is stated that the risk-weights in Basel II were calibrated by the experts with a “confidence level is fixed at 99.9%, i.e. an institution is expected to suffer losses that exceed its level of tier 1 and tier 2 capital on average once in a thousand years”

    With that let me just ask two questions

    Do you remotely believe it possible for banks to be so regulated they will only face problem “once in a thousand years” and if so what other sacrifices would humanity have to make in order for that to happen?

    Would you like to live in a world with the banks not failing and of course as a result slowly developing into having only one monstrously big and impossible to fail bank?

    Somehow regulators trying to live out their own bedroom fantasies of a world with no bank defaults completely ignored that risk-taking is the oxygen of development.

    Instead of making sure that bankers became more intelligent and more rational risk-takers they made sure it was more and more profitable for the bankers to go where the risks of default were perceived by others as low and the ratings were AAA; and which is precisely the terrain where all bank crisis in history occur… and so over the cliff of AAA rated securities collateralized by lousily awarded mortgages to the subprime sector they went… and the world with them.

  5. kiffmeister October 2, 2010 at 10:37 am - Reply

    Thanks for the comments. I guess I stand guilty as one of those who had a chance to reel in the credit rating agencies. For example, I participated in a number of structured finance-related BIS CGFS working groups, and, looking back, maybe we could have taken a harder line. However, we never found the “smoking guns” – hard evidence that there was over-reliance on ratings, and the anecdotal evidence we came across also was not conclusive.

    Here for example is a passage from the 2005 CGFS structured finance rating report:

    “Despite this “value added” by the rating agencies, market participants, in using ratings, need to be aware of their limitations. This applies, in particular, to structured finance and the fact that the one-dimensional nature of credit ratings based on expected loss or probability of default is not an adequate metric to fully gauge the riskiness of these instruments. This needs to be understood by market participants. Interviews with large institutional investors in structured finance instruments suggest that they do not rely on ratings as the sole source of information for their investment decisions, which limits the potential for misunderstood risk exposures.”

    We heard similar things in our background work for the April 2006 GFSR credit risk transfer chapter, although doubts were creeping into our thought process. That time we noted that “some questions remain as to whether all investors fully understand the risk profiles of these [structured finance] instruments, and how it differs from those of similarly rated corporate bonds.” We went on to call for differentiated ratings for structured finance ratings.

    With 20-20 hindsight there are many of us who wish we had paid more attention to and acted more vigorously on our doubts, but, as I said, we never found that “smoking gun” until it was too late. On the other hand, I’m not sure that just reducing rating reliance would have by itself averted the crisis. The models that the rating agencies used were not much different from those being used by more savvy market participants, and being batted about by academics. Of course, model herding is a whole other issue, and one that we discussed in the October 2007 GFSR.

    Anyways, again thanks for the comments, and I look forward to further dialogue!

  6. Tariq Scherer October 2, 2010 at 11:45 am - Reply

    Though I agree with the gist of this post, and in particular the over-reliance on credit ratings for risk adjusted weightings of assets, I am worried that we haven’t proposed an adequate solution to the problem just yet.

    I’ve written a bit about market price weighting measures: using stochastic variables or other market defined means for weighting collateral processes but how can we ensure that these methods are set out clearly and uniformly across markets? Also, what about assets with naturally illiquid markets and limited price discovery – are rating agencies still the key then?

    I don’t know if this is is available, but could we have a more public disclosure over ratings processes (similar to what happens with private credit records) with potentially a right of appeal over the judgement. To my knowledge, the current process is set as an independent arbiter, which in principle seems reasonable, but without external review processes don’t we risk locking ourselves into one-sided views with little scope for review? After all, an appeal towards the process could be as much due to a request for rating increase as a downgrade…

    Again though, I would have thought that market pricing would be the best forum for this: confront one opinion against the market rather than try and sway the market through an unpriced third-party…

    Tariq Scherer

  7. Per Kurowski October 3, 2010 at 2:15 pm - Reply

    Kiffmeister writes “I participated in a number of structured finance-related BIS CGFS working groups, and, looking back, maybe we could have taken a harder line. However, we never found the “smoking guns”– hard evidence that there was over-reliance on ratings”

    The primary reason for that was that you were yourselves holding the smoking gun, since what more over-reliance on ratings could there be than regulators authorizing banks to leverage 62.5 to 1, just because of some AAA ratings.

    I am no PhD, so it is hard for me to get any research noticed, but do yourselves a favor and go back and look at what happened in banking after Basel II was approved in June 2004.

  8. kiffmeister October 3, 2010 at 10:49 pm - Reply

    Per, I’m no Phd either, and I’ve never been a regulator (and neither is the IMF!), but you’re preaching to the converted. However, our work is not aimed at pointing fingers. If we were, I’m sure we would see eye to eye on many things. Now it’s time to look forward.

  9. Per Kurowski October 4, 2010 at 10:05 am - Reply

    Excuse me, but I have always considered that the IMF has functioned as the club house of the Basel Committee. If this is not the case, then it is even more important, while moving forward, for the IMF to challenge the powers of the Basel Committee; and which for all practical purposes are charging forward as if nothing has happened; now setting their eyes on controlling for cyclicality which will of course risk introducing additional dangerous distortions into the financial markets.

    This is not about pointing fingers, this is about moving forward correctly!

    If the IMF and the World Bank are not there to alert the G-20 and the rest of the world on financial regulatory failures that affect global financial stability, then who is… the duo of the not PhDs “kiffmeister & kurowski”?

  10. Vanguard October 4, 2010 at 10:11 am - Reply

    I suggest the simplest solution would be if we force credit rating agencies to publish the workings/assumptions that were used in arriving at the rating.

    The ratings are not the problem. Rather the black box wherein the data goes and comes out as a rating has been shown to be the real problem. Rating agencies had already been working with issuers on how to structure the securities to achieve the best rating (or rather game the system), why not ask them to work with investors and share with them the workings/assumptions that were used to arrive at that rating. They can charge a fees for it but then the investors can themselves verify the assumptions/formulas in the models till such time that investors become lazy and just save the models in their hard drive and take the ratings assigned by CRA.

  11. kiffmeister October 4, 2010 at 11:33 am - Reply

    We are trying to alert the world to issues we see as relevant–let me know when you have some hot tips. Anyways, as “Vanguard” says, more transparency with regards to models etc., would be helpful, although keep in mind what happened with structured finance products. In that case, model transparency may have facilitated rating shopping.

  12. Per Kurowski October 5, 2010 at 7:40 am - Reply

    kiffmeister, I am not sure whether it would classify as a “hot tip”, it shouldn´t, but since it has clearly been ignored by the regulators, allow me to bring it to the table.

    When a client, perceived as more risky, like for instance your average small business or entrepreneur, is requested to pay for instance 5 percent or more on their loans than what a triple-A rated client pays, where do you think that 5 percent or more goes to when it gets repaid? The answer is to bank equity of course. That is what we could call the market´s risk-weights.

    And when a regulator decides that a triple-A rated client generates only a 1.4 percent capital requirement for the bank (the Basel III 7 percent, adjusted by the risk-weight of 20 percent), while your average risky small business or entrepreneur generates a capital requirement of 7 percent, where do you think the about 2 percent in additional interest that your average small business or entrepreneur has to pay the bank in order to make up for the bank´s opportunity costs goes? The answer is to bank equity of course. This is what we call the regulator´s risk weights.

    And so we have a world where, out of the blue, the Basel Committee decided that all our small businesses and entrepreneurs, those whom we should be most interested that our banks finance, well they have to run with two different sets of risk-weights.

    What kind of handicap officer is the Basel Committee, taking off weights from those who have been running nicely and putting weights on those who have not run as good or are just debutants? I would have to answer that that handicap officer has completely misunderstood his role.

    The best way to end the markets’ addiction to the credit rating agencies is to end the regulator´s obsession with the credit rating agencies.

  13. kiffmeister October 5, 2010 at 1:06 pm - Reply

    Per, we have advocated that more capital be held against riskier activities, whether they be from small and medium size enterprises or large companies. The chapter suggests those than lend should take a close look at the actual credit risks and not rely exclusively on credit ratings and their associated formalized risk weights. That would be the ideal.

  14. Per Kurowski October 5, 2010 at 2:47 pm - Reply

    I agree, but only to the extent that by considering the risk of activities, do not introducing new arbitrary discrimination and distortions in the markets… I find it hard to visualize the market (and the regulators) being able to digest and measure adequately the impact of different regulatory risk-weights.

    Also never forget that while trying, desperately, to avoid the failure of banks, which can indeed have a big cost for society, you are doing so at the expense of making it more difficult for a bank to perform its role as a capital allocator, and that will have a much higher final cost for the society at large.

    There is nothing wrong with banks failing (in fact it could be nightmare if they didn’t) the problem is when they fail as a system, and that is what the regulators inadvertently promoted by developing, through the approval of minuscule capital requirements, special incentives for the banks to go where risks are perceived as low, in other words to go to the only place where there can be an accumulation of capital sufficient to bring down the system, in other words the place where AAA ratings reside.

    Being perceived as risky is weight enough. In my emerging (or perhaps now submerging country) I never witnessed a crisis because banks overinvested in something risky. When it happened it was for entirely different reasons than risk appreciation, it happened for instance because they lent money to affiliated enterprises… and which they sometimes even considered totally risk-free… until the day it was not.

  15. […] IMF makes sense October 7, 2010 // 0 In recent days the IMF has publicly argued that “markets need to end their addiction to credit ratings”. They […]

  16. […] sugerir que as ACRs sejam marginalizadas do processo de avaliação de riscos financeiros (ver aqui). Contudo, curiosamente, em nenhum dos seus aspectos, a atuação dessas agências foi considerada […]

Leave A Comment