What Would Make a Multilateral Credit Rating Agency Consequential?
Proposals for alternative sovereign credit rating agencies typically concentrate on methodology: how sovereigns should be evaluated differently, against what indicators, and over what horizon. They devote considerably less attention to a second question, of how a new evaluator becomes consequential once it exists. The Policy Profile proposing a Multilateral Credit Rating Agency, issued as part of the Roadmap for Eradicating Poverty Beyond Growth, is a case in point. It completes the first task with considerable care. It says comparatively little about the second (Muchhala and Syed, 2026).
This essay treats the Policy Profile’s proposal seriously rather than disputing it. The report starts from the proposition that prevailing approaches to sovereign credit assessment do not adequately capture long-term development, productive investment and structural transformation, and a counterweight housed within the multilateral system is a reasonable response to that pattern as identified. But methodology answers only the first of two projects that any new evaluative institution faces: changing evaluation and building coordination. This essay is concerned with the second, not how creditworthiness should be redefined, but how a redefinition comes to matter in practice, and it treats the Policy Profile as an occasion for that broader argument rather than as its subject. Although developed here through the case of sovereign credit ratings, the argument applies more broadly to new evaluative institutions seeking to reshape existing systems of governance.
A different conception of creditworthiness
The Policy Profile’s diagnosis of the existing sovereign credit rating system draws on evidence that is well established elsewhere in the literature. Textual analysis of rating committee reports finds that gross domestic product per capita, rather than measures of solvency, carries strong predictive weight in sovereign assessments, and that this weight operates over the short trajectories the industry favours (Muchhala and Syed, 2026). A related review by the United Nations Conference on Trade and Development finds that subjective judgement and sentiment enter sovereign scorecards alongside quantitative indicators in ways that create scope for systematic divergence from a country’s underlying fundamentals, even where such divergence is difficult to prove conclusively in any single case (UNCTAD, 2024). The procyclical character of rating action, in which downgrades intensify during exactly the periods when countries can least afford higher borrowing costs, has been documented across the pandemic period and beyond, with developing economies bearing a disproportionate share of downgrades relative to the severity of their economic contractions (Muchhala and Syed, 2026).
The Policy Profile’s response, a methodology built around thirty to forty year time horizons, a move away from the centrality of gross domestic product, and explicit incorporation of climate resilience and productive investment, follows from this diagnosis and holds together on its own terms. Susan Schroeder’s earlier proposals for a public credit rating agency, later extended into a fuller Multilateral Credit Rating Agency framework prepared for the Economic Commission for Latin America and the Caribbean, arrive at a similar analytical position: an alternative evaluator must differ enough from the incumbents that its approach cannot simply be absorbed into their existing methodology, or the new institution becomes redundant almost as soon as it is created (Schroeder, 2023).
This is the first project, carried out with real analytical care. It does not, by itself, resolve the second, and the remainder of this essay is concerned with why the two projects do not resolve together automatically.
Two Projects
The distinction deserves to be stated more precisely than the Policy Profile itself states it, because the wider literature on alternative sovereign credit rating agencies tends to collapse the two projects into one, treating a sound methodology as though it were already most of the work.
Changing evaluation is the work of methodology: deciding which indicators a rating should weight, over what horizon, and against what conception of a borrower’s underlying capacity. Building coordination is a different kind of work. It refers to whether the organisations that actually move money, or write the rules governing how money moves, begin building their own routine decisions around a given evaluator’s opinion rather than around an incumbent’s. An institution can complete the first kind of work in full and achieve almost none of the second, because coordination is not a property of an evaluator’s methodology. It is a property of the surrounding financial system, and specifically of how much of that system’s routine machinery is already organised around a different opinion. Stated most simply, the first project concerns knowledge. The second concerns institutions.
Bartholomew Paudyn’s account of sovereign ratings helps explain why coordination cannot simply follow once a better opinion exists. Ratings do more than describe a pre-existing fiscal reality; they participate in constructing the reality they claim to measure, in the sense developed in Callon’s work on the performativity of economic devices (Paudyn, 2014). Credit rating agencies have sustained their authority through recurrent and well documented failures, including the mispricing of structured finance instruments before the 2008 crisis and repeated procyclical misjudgements during subsequent sovereign debt episodes (Paudyn, 2014). Had coordination followed automatically from analytical accuracy, that record would have eroded the incumbents’ standing considerably more than it has. What sustains their standing instead is that decades of investor, regulator and issuer practice have already been built around their particular opinion, in the sense Dieter Kerwer describes when he treats credit rating agencies as coordination service firms whose standard of creditworthiness has become a reference point independent of any single actor’s preference (Kerwer, 2001).
This raises a definitional question the wider literature rarely poses directly. What, precisely, counts as adoption? The term could mean that an evaluator’s opinion is read and trusted, that it is cited as a reference point in due diligence, that it is written into the covenants of a bond contract, that it determines the capital an investor must hold against a position, that it is built into an internal risk model, that it enters the rule governing which securities a benchmark index may include, or that it becomes one input among several in a regulator’s own prudential formula. These are not the same achievement, and an evaluator can secure some of them without securing others. An agency can be widely read without a single one of its opinions ever entering a regulatory formula, and a rating can be embedded in regulation without investors placing any independent trust in its accuracy. For the purposes of this essay, adoption means the point at which an organisation begins to build its own routine practice, whether a contract, a model, a mandate, or a regulatory formula, around a given evaluator’s judgement, such that removing that judgement would require the organisation to change its own practice rather than simply substitute one number for another. That is a considerably higher bar than being read, and it is the bar the Policy Profile’s proposed institution would eventually need to clear.
Establishment, recognition and adoption
Three stages describe the distance between a design on paper and an evaluator whose judgement organisations have built into their own practice: establishment, recognition, and adoption. Establishment is the first project made institutional: governance, funding, methodology, personnel, an institutional home, and the data and analytical capability to produce opinions at all. This is what the Policy Profile addresses directly, proposing a Multilateral Credit Rating Agency with independent governance, methodological distinctiveness, and access to the United Nations system’s own data and research capacity (Muchhala and Syed, 2026).
Recognition is the first moment of the second project, and it means formal acknowledgement of an evaluator’s standing by the bodies that sit between the evaluator and the market: securities regulators, central banks, prudential supervisors, the International Organization of Securities Commissions, the Basel Committee, multilateral development banks, and the professional associations through which analysts and portfolio managers coordinate their own practice. Recognition is necessary but not sufficient, because it grants an evaluator standing to be used without compelling anyone to use it.
Adoption is the second and decisive moment of the second project, defined above: the point at which banks, pension funds, asset managers, sovereign debt offices, the law firms that draft bond covenants, the index providers who decide which securities a benchmark may hold, the risk managers who calibrate internal models, and the data vendors who distribute ratings into trading systems all begin building their own routine practice around a given evaluator’s opinion. Adoption, not recognition, is what would allow a Multilateral Credit Rating Agency’s judgement to move the price sovereigns pay to borrow, and the history of previous attempts to diversify the sovereign credit rating market shows how far recognition can travel without adoption following behind it.
The United States Securities and Exchange Commission’s Nationally Recognised Statistical Rating Organisation designation illustrates the gap between recognition and adoption with particular clarity. Introduced (promulgated) in 1975, the designation identified which agencies' ratings could be relied upon for specific regulatory purposes. Although additional agencies gradually obtained NRSRO status over subsequent decades, market activity remained overwhelmingly concentrated among Moody’s, Standard & Poor’s and Fitch, illustrating that formal recognition alone did not produce widespread market adoption (U.S. Securities and Exchange Commission, 2003). Recognition under this regime was explicitly tied to market use rather than independent of it. The Commission’s own criteria centred on whether an agency was already recognised as credible by the predominant users of ratings, meaning that formal recognition depended on the market having already adopted an agency’s opinions (Kerwer, 2001). New entrants faced something close to a regulatory paradox: without the designation they struggled to attract the investor base needed to be considered widely used, and without that investor base they could not obtain the designation (Kerwer, 2001).
A comparable pattern appears in Europe, and it separates recognition from adoption just as clearly. The European Union’s Credit Rating Agency Regulation created a formal registration and certification process explicitly intended to widen the field beyond the three dominant firms, and by the mid-2010s more than two dozen agencies had registered or been certified across the Union (European Securities and Markets Authority, 2015). Registration is recognition, granted and achieved. A 2021 study using the European Securities and Markets Authority’s own dataset of ratings issued since 2015 found that the three largest credit rating agencies nonetheless retained an aggregate market share exceeding ninety per cent across the European Union, and that small credit rating agencies operating in local, single-rating markets remained locked out of the larger contracts in which multiple ratings are solicited for the same issuer or instrument (European Securities and Markets Authority, 2021). The concentration levels involved, measured using the Herfindahl-Hirschman Index common in competition analysis, sat well above the threshold generally understood to signal a highly concentrated market (European Securities and Markets Authority, 2021). Recognition had travelled. Adoption had not followed it.
The Financial Stability Board’s own Principles for Reducing Reliance on Credit Rating Agency Ratings offer the mirror image of the same gap: rather than helping new evaluators gain adoption, the Board tried to strip mechanistic adoption away from the incumbents by regulatory instruction, following the 2012 Roadmap endorsed by the Group of Twenty. Even with the weight of coordinated regulatory reform behind it, the Board’s 2014 peer review found that most jurisdictions had made only slow progress developing alternative standards of creditworthiness, and that the most substantial completed reform, under the Dodd-Frank Act in the United States, still left many financial institutions dependent on the same rating references embedded in prudential capital frameworks (Financial Stability Board, 2014). Subsequent revisions to the Basel Framework continued to permit external credit ratings to determine risk weights for sovereign exposures, provided the issuing agency met certain recognised standards (United Nations, Financing for Sustainable Development Office, 2022). If adoption resists deliberate regulatory instruction to remove it, it should not be expected to arrive automatically once a new evaluator is established and recognised.
Adoption is Cumulative
No single obstacle explains why adoption, once granted to Moody’s, S&P Global, and Fitch, has proven so difficult to redirect toward any alternative evaluator, however well-established or recognised. The reason is that adoption, once achieved, is not held by any one institution. It is distributed across a dense and mutually reinforcing set of practices that together function as something closer to an operating infrastructure than to a market position a competitor might simply outbid.
Consider what that infrastructure actually contains. Herwig and Patricia Langohr’s account, presented to the Organisation for Economic Co-operation and Development’s Competition Committee, identifies three functions a sovereign credit rating performs at once: it resolves information asymmetry between borrower and lender, it provides a consistent scale for comparison across an entire portfolio of holdings, and it offers a contractible language that can be written directly into private contracts and regulations (Organisation for Economic Co-operation and Development, 2010). Each function depends on the others already being in place across the market, which is why an investor who has built a portfolio around one agency’s comparative scale has little reason to duplicate that effort for a second, unfamiliar scale, and why an issuer with a working relationship with one or two established credit rating agencies has little reason to divide scarce management time cultivating a third.
The routine embedding of ratings into capital adequacy formulae, investment mandates, and disclosure exemptions extends this infrastructure into regulation itself, transforming a voluntary market opinion into something closer to a rule, without a matching increase in the accountability of the agencies producing it (Kerwer, 2001). Benchmark construction extends it further still. Bond indices built around an investment grade threshold force mechanical selling once an issuer crosses that line, regardless of whether the underlying agency’s judgement was well founded, embedding the incumbents’ scale into portfolio rules that many investors do not set themselves and cannot easily override. Internal risk models, calibrated over years against the incumbents’ historical default data, extend it again, since building an equivalent model around an unfamiliar evaluator’s opinion requires its own multi-year track record before a risk manager can defend using it in place of the model already in use. The empirical evidence on market power confirms that this infrastructure reinforces itself over time rather than eroding. Research using a global sample of corporate credit ratings across twenty-seven developed markets finds that greater market share for an incumbent evaluator is associated with tighter, more conservative rating standards, and that this pattern reverses only once a local competitor gains equivalent regulatory recognition (Hung, Kraft, Wang and Yu, 2022).
None of these practices, taken alone, would be difficult to dislodge. A single bond covenant can be redrafted. A single risk model can be recalibrated. A single benchmark rule can be amended. What makes adoption difficult is that dozens of such practices, spread across law firms, index providers, risk managers, data vendors, and regulators who rarely coordinate with one another directly, already point toward the same three evaluators, and changing any one of them in isolation accomplishes very little, since the rest of the infrastructure continues to organise itself around the incumbents regardless. Adoption, in other words, is cumulative. It cannot be granted by a single regulatory act or won through a single demonstration of superior methodology, because no single actor holds enough of the infrastructure to grant it alone. Adoption, described this way, resembles an ecosystem rather than a market decision.
Lessons from attempts to diversify the market
The empirical record of previous efforts to build alternatives to the three dominant credit rating agencies is more informative than a straightforward story of failure would suggest, once it is read against the three stages rather than as a single undifferentiated history.
Proposals for a European public credit rating agency, prompted directly by the eurozone sovereign debt crisis, did not clear even the first stage. They generated considerable political enthusiasm in the early 2010s before stalling over concerns about credibility, financing, and the risk that a publicly backed evaluator would be accused of market manipulation, and the European Commission subsequently concluded that a new European agency would add little additional information for investors already served by the incumbents (Scheinert, 2016). No institution was ever fully established, so the later stages were never reached. Smaller registered European agencies such as Scope, Creditreform and Axesor tell a different story: they achieved both establishment and recognition under the Credit Rating Agency Regulation, yet the 2021 European Securities and Markets Authority analysis found their combined share of the broader, cross-border market for sovereign and major corporate ratings remained marginal relative to the three dominant firms, even a decade after registration became available to them (European Securities and Markets Authority, 2021). Establishment and recognition were both achieved. Adoption was not.
The African Credit Rating Agency offers a case still in the establishment stage, and one that illustrates a further obstacle that can intervene before recognition is ever tested: the incumbents’ capacity to absorb a potential alternative before it completes the pathway on its own terms. Moody’s acquired a majority shareholding in Global Credit Rating, the largest agency headquartered in Africa, in 2022, and holds stakes in the Middle East Rating agency and the West African Rating Agency as well (African Peer Review Mechanism, 2024). S&P Global has only recently acquired Agusto & Co. Each acquisition removes a domestically rooted evaluator from the pool of potential alternatives and folds its analytical capacity into one of the very firms the alternative was meant to counterbalance, intercepting the pathway from establishment to adoption rather than contesting it. A market structure open enough to permit new entry is not the same as a market structure that allows new entrants to grow into genuine substitutes for the incumbents, and acquisition is as effective a barrier to that growth as exclusion would be.
China’s Dagong Global Credit Rating Company and the certification of firms such as Egan-Jones in the United States extended the roster of recognised alternatives further still, and yet the aggregate pattern across these episodes is consistent: establishment has proven achievable, recognition has proven achievable, and adoption at the scale needed to move sovereign borrowing costs has not followed either (European Securities and Markets Authority, 2015). The lesson is not that alternative agencies have failed as institutions. Most continue to operate, several profitably, within the niches they have secured. The lesson is that establishment and recognition, even sustained for a decade or more, do not on their own produce adoption, and a strategy for a Multilateral Credit Rating Agency needs to treat adoption as a separate undertaking rather than an eventual by-product of the first two stages.
Building Adoption
It is worth pausing on why design debates about new evaluative institutions consistently privilege methodology over adoption, since the Policy Profile is far from alone in this respect. Part of the reason is practical. A methodology can be authored by a small team working from existing data and published critique, and its coherence can be judged by reading the document itself. Coordination cannot be authored in the same way. It requires persuading regulators, index providers, risk managers and law firms who owe the new institution nothing, over a period measured in years rather than months, and its success cannot be verified by reading anything the institution produces. Institutional design is, in this sense, easier to imagine than institutional change, and proposals for new evaluators understandably gravitate toward the part of the problem that can be settled on paper. The omission is not a flaw particular to this Policy Profile. It is closer to a structural feature of how such proposals get written, which is one further reason the second project deserves the same degree of explicit attention as the first.
Susan Schroeder’s own proposals for a public and then multilateral credit rating agency addressed the adoption problem more directly than most of the surrounding literature, and her recommendations point toward what a deliberate adoption strategy for the Policy Profile’s proposed institution might include. She argued that a new agency would need to establish a documented track record validating its assessments against the accuracy of private ratings before governments could reasonably be expected to write its opinions into regulation, and suggested the agency actively pursue Nationally Recognised Statistical Rating Organisation status in the United States specifically in order to make its judgements usable for regulatory purposes rather than informational ones alone (Schroeder, 2013; Schroeder, 2015). This is a strategy aimed squarely at recognition, built on the understanding that recognition itself has to be pursued rather than assumed. She further proposed a subscription-based revenue model, drawing on the historical precedent of credit rating agencies before the shift to the issuer-pays structure, as a route to the kind of institutional independence and staffing scale that sustained credibility over decades has typically required (Schroeder, 2023).
A more recent proposal from the United Nations Conference on Trade and Development takes a different route to the same underlying problem, and the contrast is instructive. Rather than positioning a new evaluator to compete directly with the incumbents across their existing client base, where the infrastructure described above is already fully built, the 2024 review proposes a technical assistance process aimed specifically at the roughly forty developing countries that currently have no sovereign rating at all, reasoning that a new institution’s judgement is more likely to be incorporated into financing decisions where no competing incumbent opinion, and no competing infrastructure, already occupies that space (UNCTAD, 2024). This is not necessarily a stronger design than the Multilateral Credit Rating Agency proposal, and the Policy Profile’s ambition to challenge the incumbents’ assessment of already-rated sovereigns has its own case to make. It does suggest that adoption is considerably easier to secure where an evaluator enters a genuine gap in coverage than where it must displace an infrastructure that investors, regulators and issuers have organised decades of practice around.
This points to a question the Policy Profile does not yet ask directly. Adoption almost certainly cannot happen everywhere at once, and the sequencing of early adopters may matter as much as the design of the institution itself. Daniel Berliner and Aseem Prakash’s study of how private governance standards spread among firms finds that adoption depends on the surrounding institutional context of the adopter, not merely on pressure from the standard-setter; organisations already embedded in a compatible institutional environment adopt new standards more readily than those facing an indifferent or hostile one, even when the standard’s own design is unchanged (Berliner and Prakash, 2014). Applied to a Multilateral Credit Rating Agency, this suggests that the plausible first adopters are not private asset managers or index providers, whose own practice is already organised around the incumbents’ infrastructure, but organisations institutionally closest to the agency’s own sponsors: multilateral development banks and United Nations agencies already committed to the analytical premises the Policy Profile sets out, together with regional development banks, guarantee agencies, and export credit agencies whose mandates already favour long-horizon development lending over short-term portfolio considerations. A demonstrated track record among this narrower set of early adopters, whose institutional proximity to the agency’s own mission lowers the cost of adoption considerably, would then supply the evidentiary basis on which broader regulatory recognition, and eventually adoption among private asset managers and index providers, could be sought. Attempting to secure adoption from the private financial infrastructure first, before this narrower base exists, risks repeating the pattern of the European public credit rating agency proposal, which found no natural constituency of users once political enthusiasm faded (Scheinert, 2016).
The Multilateral Credit Rating Agency proposal already contains several of the elements a full strategy would need: institutional independence, transparent governance, methodological credibility distinct enough to avoid absorption by the incumbents, and access to the United Nations system’s own data and research capacity (Muchhala and Syed, 2026). What the proposal would gain from making explicit is that these elements complete only the first project and the first of the three stages, and that establishment must be followed by a deliberate strategy for recognition, sequenced deliberately through the organisations institutionally closest to its own mission, before adoption among the wider financial infrastructure becomes a realistic aim.
The Policy Profile identifies that the existing system for assessing sovereign creditworthiness deserves reconsideration, and its proposal for a Multilateral Credit Rating Agency completes the first project, the case for evaluating sovereigns differently, with real care. The second project, building coordination, remains open, and this essay has argued that it should be treated as a distinct undertaking, with its own definition of success and its own sequencing, rather than as a task that resolves itself once establishment is complete.
The record examined here, of registration without market share, of certification without adoption, of incumbents absorbing the very alternatives meant to challenge them, points to a single underlying claim. The three dominant credit rating agencies did not become consequential because they produced more persuasive evaluations than their rivals. They became consequential because successive generations of banks, regulators, index providers and law firms incorporated those evaluations into their own routines, one contract and one model at a time, until removing them would have meant rebuilding the routine itself. The same logic runs forward as well as backward, and it extends well beyond credit rating agencies to any new evaluative institution entering an existing system of governance. Institutions do not become influential simply because they are created, or even because they are formally recognised. They become influential at the point where other organisations begin organising their own contracts, models, and mandates around them, and that point has to be built, deliberately and in a particular sequence, rather than assumed to follow from good design. Building it is a longer project than building the institution itself, and it is the project on which the case for a Multilateral Credit Rating Agency, like the case for any new evaluative institution, will ultimately be judged.
References
African Peer Review Mechanism (2024) An Africa Credit Rating Agency (AfCRA): Key Shaping the New Global Financial Architecture. African Peer Review Mechanism.
Berliner, D. and Prakash, A. (2014) ‘Public authority and private rules: how domestic regulatory institutions shape the adoption of global private regimes’, International Studies Quarterly, 58(4), pp. 793–803.
European Securities and Markets Authority (2015) Technical Advice: Competition, Choice and Conflicts of Interest in the Credit Rating Industry. ESMA/2015/1472.
European Securities and Markets Authority (2021) ‘The market for small credit rating agencies in the EU’, in ESMA Report on Trends, Risks and Vulnerabilities, No. 2, 2021, pp. 82–93.
Financial Stability Board (2014) Thematic Review on FSB Principles for Reducing Reliance on CRA Ratings: Peer Review Report, 12 May 2014.
Hung, M., Kraft, P., Wang, S. and Yu, G. (2022) ‘Market power and credit rating standards: global evidence’, Journal of Accounting and Economics, 73.
Kerwer, D. (2001) Standardising as Governance: The Case of Credit Rating Agencies. Preprints aus der Max-Planck-Projektgruppe Recht der Gemeinschaftsgüter, Bonn, 2001/3.
Muchhala, B. and Syed, M. (2026) ‘Policy Profile 5.2: Curbing the dominance of private credit ratings agencies through multilateral credit rating’, in De Schutter, O., The Roadmap for Eradicating Poverty Beyond Growth. New Economies Eradicating Poverty.
Organisation for Economic Co-operation and Development (2010) Competition and Credit Rating Agencies. Competition Committee Roundtables on Competition Policy, No. 152, DAF/COMP(2010)29.
Paudyn, B. (2014) Credit Ratings and Sovereign Debt: The Political Economy of Creditworthiness through Risk and Uncertainty. Basingstoke: Palgrave Macmillan.
Scheinert, C. (2016) The Case for a European Public Credit Rating Agency. European Parliamentary Research Service Briefing, PE 589.865.
Schroeder, S. (2013) ‘A template for a public credit rating agency’, Journal of Economic Issues, 47(2), pp. 343–350.
Schroeder, S. (2015) Public Credit Rating Agencies: Increasing Capital Investment and Lending Stability in Volatile Markets. New York: Palgrave Macmillan.
Schroeder, S. (2023) ‘Multilateral Credit Rating Agency’, in Pérez Caldentey, E. and Villarreal, F. G. (eds) Innovative Financing Instruments in Latin America and the Caribbean. Santiago: Economic Commission for Latin America and the Caribbean.
United Nations, Financing for Sustainable Development Office (2022) Credit Rating Agencies and Sovereign Debt: Challenges and Solutions.
UNCTAD (United Nations Conference on Trade and Development) (2024) Credit Rating Agencies, Developing Countries and Bias. Geneva: UNCTAD.
U.S. Securities and Exchange Commission (2003) Report on the Role and Function of Credit Rating Agencies in the Operation of the Securities Markets.