> ## Content Index
> Fetch the complete content index at: https://www.drdanielcash.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# PUBLIC - When Reform Ambition Outruns Institutional Reality
- URL: https://www.drdanielcash.com/public-when-reform-ambition-outruns-institutional-reality/
- Published: 2026-11-18T09:54:38.000Z
- Updated: 2026-11-18T09:54:38.000Z
- Author: Daniel Cash
- Tags: Development Finance, Economic Governance, Insights

Advocacy regarding the role credit rating agencies may play in relation to development has ramped up in recent years with the multilateral sector turning its attention to the issue of ‘financing for development’. One critical element of this infrastructure is civil society, which exists to observe, push, pull, and overall present a system of checks and balances on behalf of those who may need a voice in such circumstances. To that end, a recently published C20 Policy Pack 2026 asks G20 governments to lower borrowing costs for the Global South and, as part of that effort, to support a review of the dominant role of private credit rating agencies in favour of a multilateral initiative (C20, 2026). Behind that one clause sits a more developed case. Bhumika Muchhala and Maria Syed’s report for Third World Network (TWN) sets out objectives, analytical foundations and a methodology for a Multilateral Credit Rating Agency, or MCRA (Muchhala and Syed, 2026). Neither document records any institutional commitment to fund, house or govern such an agency but forcefully advocates for it.

I have spent much of my career arguing that sovereign credit rating governance requires reform, and I share the frustration behind these documents. Market concentration, with three firms issuing over 94 per cent of sovereign ratings on TWN’s account, the issuer-pays conflict and procyclical downgrades all very much deserve the attention civil society has given them (Muchhala and Syed, 2026). My objection lies elsewhere. The proposals move from an accurate account of these problems directly to a new international institution, without demonstrating that the institution would solve them or, for a lot of the time, the implications that such an organisation and its creation would produce, both positively and negatively. I am of the strongest opinion that these are critical ingredients in the recipe of reform in order to garner the authority needed to undertake such a gargantuan effort in evolving the international financial architecture.

## Sign up for Dr Daniel Cash

Research, thought, and advisory

Subscribe 

Email sent! Check your inbox to complete your signup. 

No spam. Unsubscribe anytime.

# Three ambitions assigned to one institution

Read together, the two documents ask the proposed agency to reduce sovereign borrowing costs, to give greater weight to development and long-term potential, and to break the dominance of the incumbents. Each requires its own mechanism, and they need not reinforce one another.

A credit rating agency does not set the price of sovereign debt. They are a critical component of the wider ecosystem yes, but setting the price – seemingly the most pertinent issue for advocates – sits elsewhere in the chain, most noticeably in the relationship between investment bank as bond arranger and the state as their client but, in reality, across a broad range of actors in the investment space. A credit rating agency’s judgements can move financing conditions, but the investment ecosystem sets the price, and the advocates’ own evidence shows this (I would also urge readers to read this helpful World Bank piece for a [step-by-step guide](https://documents1.worldbank.org/curated/en/491301554821864140/pdf/Issuing-International-Bonds-A-Guidance-Note.pdf?ref=drdanielcash.com) to the Eurobond issuance process). In the Latindadd case studies, Argentine spreads widened after the August 2019 primaries and before the formal downgrades, the market response to Zambia’s 2020 default and rating actions was muted, and Colombian spreads ended the month after the June 2025 downgrade below their starting level (Muchhala et al., 2026). The authors conclude that ratings validate and amplify market stress, and I think they are right. An amplifier can make conditions worse, but it does not follow that a second agency with different opinions would make them better, since the price would still be set by an investment ecosystem which organise their decisions around the traditional ratings from the established agencies.

The second ambition carries a different tension. The TWN report holds that public and climate investment improve creditworthiness over time, and that the new agency would treat welfare-improving expenditure accordingly (Muchhala and Syed, 2026). Often that will be true. Yet a sovereign can pursue a socially valuable investment programme while its near-term capacity to service debt deteriorates, and an independent evaluator must be prepared to say so, even against its founding purposes. The report also proposes a temporary moratorium on ratings during exogenous shocks. Suspending rating actions does not predetermine an assessment, but it withholds material credit information at the moment investors most need it, and an evaluator that falls silent under stress gives them little reason to consult it at any other time. It may not necessarily ‘withhold’ information at the critical time because an agency can still produce research instead of a rating, but the timeliness of such an intervention then becomes important, and the market may very well read into the research the same way they read into a rating. This raises the question of what exactly the institution advocates are pushing for is in reality. Can an agency be a *credit rating* agency and a *development rating* agency at the same time? I will come back to this later.

The third ambition demands far more than the creation of a competitor. A few months ago, I [examined a related problem](https://www.drdanielcash.com/what-would-make-a-multilateral-credit-rating-agency-consequential/): the considerable distance between establishing a new evaluator and making its judgements consequential. I distinguished establishment, recognition and adoption, and argued that the authority of the dominant agencies rests on investors, regulators, index providers, law firms and risk managers who have built their contracts, models, and mandates around those agencies’ opinions. That argument concerned what happens after an institution exists. This one concerns the responsibility of advocating its creation.

# What a credible evaluator costs

The TWN report contains no costing. Its discussion of obstacles is confined to the structural power of the financial sector, the gatekeeping role of the incumbents and their epistemic authority (Muchhala and Syed, 2026). These obstacles are real, but none concerns what it would take to build and run the agency: specialist analysts, data infrastructure, rating committees, compliance, regulatory recognition and a funding base that outlasts the first budget cycle, all of which are continuing obligations. It also does not discuss the appetite of those who would fund it, and for how long, and under what conditions.

The most detailed figures around what a public agency may need come from Susan Schroeder, whose academic work underpins much of the current advocacy. She estimates a skeletal structure of 32 analysts at US$6.5 to 7.1 million and US$24 to 26 million a year at full scale (Schroeder, 2023). Sovereigns and central banks would fund the start, with subscriptions taking over. To her credit, she costs her design, but it revives the subscription structure the industry abandoned because, as she acknowledges, revenues could not keep pace with costs. She also reports that the Bertelsmann Foundation’s proposed non-profit agency failed because sufficient funds could not be gathered or maintained (Schroeder, 2023), an entity I reviewed in my [very first academic article](https://papers.ssrn.com/sol3/papers.cfm?abstract%5Fid=2800725&ref=drdanielcash.com) back in 2016\. Neither the C20 nor the TWN report engages with these figures or that precedent. Their silence does not prove that an MCRA is unaffordable, and Schroeder’s estimates describe her own model, which need not match the agency now advocated. What it does establish is that financial feasibility remains undemonstrated in the documents asking governments to act.

Launching an institution and sustaining a credible evaluator are different financial propositions. Advocates should be able to say what the agency costs in its fifth year as well as its first, how many sovereigns a given budget allows it to rate with the depth its methodology demands, who meets the shortfall if subscriptions disappoint, and what remains when the political enthusiasm that launched it moves on. It may very well be argued by advocates that this is not their role, and others would need to design and understand the feasibility in more distinct detail, which may be a fair response; the question then would be to whom does that responsibility fall? The thirty-to-forty-year horizon proposed in the TWN report compounds the difficulty, because it promises analysis whose accuracy cannot be demonstrated for decades. Long-term horizon models can be evaluated through historic testing or intermediate outcomes, but the uncertainty within credit analysis is naturally increased by expanding the time horizon. The African Credit Rating Agency, a separate continental initiative with its own design, will be instructive on these questions, but it should not be read as a pilot for the MCRA, for many reasons but chiefly because it is a private-led endeavour by design despite its label and conceiving at the hands of the African Union.

# The multilateral label

The word multilateral is doing considerable work in these proposals, and it should not be mistaken for independence, credibility, or accountability. An agency financed by sovereigns and central banks escapes the issuer-pays conflict only to acquire others. Schroeder recognises the risk and recommends keeping national governments off the advisory committee (Schroeder, 2023), but advisory seats matter less than budgets. The proposals do not say what happens when the agency concludes that a sponsoring government warrants a downgrade, or when its sponsors’ development preferences diverge from its assessment of default risk.

The mandate sharpens the difficulty. The TWN report charges the agency with producing ‘fairer risk assessments’ for developing countries (Muchhala and Syed, 2026), and the C20 places the initiative beneath a recommendation to lower borrowing costs. A fairer assessment need not be a more favourable one. An institution founded on that pairing will nonetheless be expected, by its sponsors and by the markets pricing its output, to deliver ‘better’ ratings, and without explicit safeguards it risks having part of its conclusion written into its constitution. This also raises a pertinent question that a new agency would have to answer (and one which AfCRA will have to answer very shortly) – what does ‘fairer’ mean? If it incorporates different data sets, then how and why were those selected? What about them makes it ‘fairer’? What if such data sets cannot be captured well enough, will they be partially included or abandoned? I spoke with somebody recently about the issue of gender and the experiences of women on the African Continent – particularly pertinent issues – and their relation to the informality of African business output: how would/should this be injected? Can it be reliably injected at all? The issue here, to be clear, is not about whether data can be collected (itself a pertinent question), but instead on the credit relevance of the data that is collected. Schroeder adds further roles, envisaging an agency that presses for moratoriums on downgrades and houses a fund to purchase distressed sovereign debt (Schroeder, 2023). An institution that rates sovereigns while holding their debt faces conflicts at least as serious as those it was created to remedy. These additional roles arguably alter the very nature of the proposed *credit rating* agency and morph it into something else entirely. If that is the design, what would private investors make of it?

Civil society rightly demands accountability from the incumbents, and the TWN report commits its agency to independence and methodological transparency. An early-stage proposal need not contain a finished constitution, I concede. It should, however, explain how independence would be protected at the point where political or developmental objectives collide with a credit judgement, because that is where an evaluator’s credibility is decided. The documents assert independence without showing how it would survive that collision, and the standards applied to Moody’s, S&P Global, and Fitch should apply with equal force to their proposed counterweight.

# The opportunity cost of institutional ambition

Development finance is contracting, and the same C20 Policy Pack reports that official development assistance fell by over 23 per cent in 2025 and is projected to fall a further 6.9 per cent in 2026 (C20, 2026). Another international evaluator would draw on money, on a small pool of sovereign specialists, on political capital and on years of sustained attention, and each of those resources potentially has other uses. This is an uncomfortable conversation that is often not undertaken when new initiatives are proposed – are we complemented, adding to, cannibalising, or distracting? If an initiative or organisation truly embodied this conversation and concluded that they would potentially be doing the latter two, would they stop? This is not to suggest anything nefarious, but the realities of reform and funding, in relation to the pressures of securing funding (quite often for organisational survival) and, then, once funding is secured whether a process can be stopped or amended fully.

The attraction of a new institution is understandable. Advocates who have watched the dominance of the incumbents survive more than a decade of post-crisis reform can reasonably conclude that the existing architecture offers no workable route, and a new agency supplies what incremental reform does not: a visible response, a focal point for mobilisation, and something governments can announce as progress. Those are political virtues, and they are separate from operational effectiveness. The perception that no alternative exists is also contestable.

Several alternatives act more directly on the problems the advocates identify. Weak sovereign data systems, which feed the discretionary judgements critics attack, can be and are being strengthened. Debt management offices can be and are being equipped to engage the existing agencies on equal technical terms, and targeted regulatory and contractual reforms can and are beginning to reduce the mechanical reliance on ratings that converts opinions into forced selling. Such reform does move: in May 2026, Fitch amended its criteria to permit temporary payment pauses without a default classification, a narrow change hedged with conditions but a real one (Muchhala and Syed, 2026), and the Sevilla Commitment laid the basis for a recurring high-level meeting on credit ratings within the Economic and Social Council. None of these routes is guaranteed to work or necessarily be more cost effective. Institutional reconstruction should nonetheless be judged against them, and advocates of an MCRA should explain why their proposal is a better use of scarce resources than interventions acting more directly on the problems they describe.

Civil society performs an essential function in challenging entrenched interests and widening the range of policies governments will consider, but proposing major changes to international financial governance carries matching responsibilities. Advocates should identify the precise problem, demonstrate the mechanism through which their intervention would address it, provide a credible institutional and financial model (or at least commission or encourage others to do it for them), confront the conflicts their design creates, and show why it deserves priority over other uses of the same resources. The current proposals go some way towards the first of these tests and fall short of the rest. A multilateral agency might yet prove workable, but the present documents have not shown that it would.

I do not accept that the ambition to reconstruct an international institution is inherently more serious than the slower work of improving the architecture that exists. The persistence of the problems explains why reformers reach for a new institution and tells us nothing about whether it would work. The urgency of reform is not evidence of the feasibility of a particular reform proposal, and the distinction carries consequences beyond this one agency. Money, institutional attention, and political capacity are finite. When they are committed to a poorly specified intervention, the cost falls on the reforms left unpursued, and it is paid whether or not the institution is ever built.

# References

C20 (2026) *C20 Policy Pack 2026*. Civil 20\. Available at: www.civil20international.org.

Muchhala, B. and Syed, M. (2026) *Curbing the Dominance of Private Credit Rating Agencies Through Multilateral Credit Rating*. TWN Global Economy Series 34\. Penang: Third World Network.

Muchhala, B., Luna, J., Syed, M. and Miranda, P. (2026) *Dancing to Their Tune: Credit Rating Agencies, Sovereign Risk, and Financing Conditions in the Global South*. Lima: Latindadd.

Schroeder, S. K. (2023) ‘A 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, pp. 155-183.