Co-sponsors: Center for Economic and Policy Research, ActionAid USA, MENAFem, Bretton Woods Project, Latindadd

Moderator:

• Niranjali Amerasinghe, Executive Director, ActionAid USA

Panellists:

• Andrés Arauz, Senior Research Fellow, CEPR
• Maia Colodenco, Director of the Global Initiatives Division of Suramericana Vision
• Jon Sward, Environment Manager, Bretton Woods Project

Video of the event is available on the IMF’s website.

Niranjali: For those unfamiliar, SDRs – Special Drawing Rights – are reserve assets used in times of economic crisis. There have been growing calls to both issue new SDRs and reform the system, as it’s currently neither equitable nor predictable. SDRs are useful in crises, and we’re in a moment of deep and overlapping global crises, with many countries already in debt distress well before recent developments. This is a pressing issue particularly for countries in the Global South.

The 2021 SDR issuance during the COVID-19 crisis demonstrated their value, especially for low-income countries. SDRs enabled governments to balance budgets, expand fiscal space, and procure essential goods like medical supplies. We must preserve SDRs as part of the international crisis response toolkit.

ActionAid has previewed a report (due out in May) that highlights the human cost when governments, facing debt distress, cut public services. These cuts have serious implications for people living in poverty and marginalisation. CEPR also has distributed useful materials outside. Today’s focus isn’t a broad discussion on SDRs, but a technical dive into accounting rules that limit their use. These are incremental but impactful changes that can be made within the IMF without major political hurdles. We’ll look at how SDRs are accounted for, how they appear on balance sheets, and why this matters.

One note: we had planned to include an IMF representative, who unfortunately pulled out just a few days ago.

Andrés: What can we do now, and what’s coming? The key takeaway on SDR accounting rules is this: IMF recommendations are not legally binding. Countries are not obligated to follow them. The IMF itself stated in its 2021 guidance, following the major SDR allocation, that the Articles of Agreement do not prescribe a specific accounting treatment for SDR allocations. It further notes that countries enjoy significant flexibility in managing SDRs – including using them directly for budget support.

Crucially, this guidance confirms governments can use SDRs for health, climate, hunger, or development purposes, and can record them on the government’s balance sheet instead of the central bank’s, depending on domestic legislation. The IMF’s Government Financial Statistics Manual (2022) reinforces this, noting that in some jurisdictions (e.g. the EU), SDRs may not be considered debt. This matters: if not classified as debt, SDRs can be more easily deployed, rechannelled or donated.

The Balance of Payments Manual (BPM) also affirms that its statistical recommendations are not legal rules – leaving room for national discretion. Countries can choose to record SDRs as either government or central bank assets and classify them as equity rather than debt.

Now, how did SDRs evolve from being “paper gold” to being treated as perpetual foreign debt?

The historical record shows a shift:

• 1977 (BPM4): SDRs were seen as equivalent to gold monetisation – created without exchange or transfer. They were conceived to provide global reserve liquidity without imposing costs on other countries.
• 1993 (BPM5): Reiterated that SDRs have no liability, akin to monetary gold. There was no obligation to repay allocations – no counterpart debt.
• Early 2000s: All IMF manuals (BPM, Government Finance Statistics, and Monetary Financial Statistics) were consistent: SDRs were not liabilities, and could be recorded on the government’s balance sheet when fulfilling monetary functions. The Monetary Statistics Manual even classified SDRs under equity.

But by 2004, this view began to shift. IMF staff and statisticians started revising definitions. In 2006, the Balance of Payments Committee – following IMF staff recommendations – reclassified SDR allocations as liabilities, despite lack of consensus in the technical group.

This culminated in the publication of BPM6 (2008), which stated that SDR allocations constitute debt. The justification? A change in the definition of debt – previously defined as requiring payment of principal and interest, now changed to principal or interest. Since SDR usage involves small interest payments, it was now deemed to meet the debt criteria.

This change impacted debt statistics, accounting, and debt sustainability analyses (DSAs). While BPM6 prescribed recording SDRs as gross liabilities, this conflicted with the IMF’s own SDR Department, which handles them differently.

By 2014, other manuals (e.g. Government Finance Statistics) followed suit, calling SDR allocations “debt,” though still allowing countries to choose how to record them domestically.

In practice, when the 2009 SDR allocation happened, most countries were still using BPM5, recording SDRs as equity. As late as 2015, 83 countries still followed BPM5. But by the time of the 2021 issuance, most had adopted BPM6 – classifying SDRs as debt. Unsurprisingly, these countries were also the ones most actively using SDRs post-allocation.

The problem today is that the IMF’s position is inconsistent: it advises classifying SDRs as debt for statistical purposes, but not for DSAs – the metrics used in real-world debt decision-making. This contradiction presents a critical opportunity to push back.

So, what can we do?

The upcoming BPM7 presents a chance to influence changes to the accounting rules. It will include updated treatment of SDRs. Civil society, think tanks, and IMF member countries should mobilise to push the Fund’s statisticians to restore the equity classification of SDRs, or at least introduce greater flexibility. Doing so would facilitate rechannelling and donation of SDRs, aligning with their original purpose as a cost-free global liquidity tool.

Maia: I’m glad to be here to discuss Argentina’s experience with SDRs and how to maximise their impact. The last IMF SDR allocation was in 2021, distributing a historic $650 billion to help countries during Covid-19. However, allocations are based on IMF quota shares, not vulnerability, which calls for reform. Argentina received $4.3 billion (3.05 billion SDRs) amid economic hardship and large IMF debt repayments from a $4 billion stand-by agreement. We used our SDRs to repay the IMF, which increased fiscal space and eased financial pressure.

The IMF’s Articles of Agreement don’t specify which government body manages SDRs – it’s decided nationally. Most countries’ central banks hold SDRs as assets and liabilities, boosting reserves. Alternatively, treasuries record them as financial assets. In Argentina, the Ministry of Finance received SDRs, registered at the central bank. The central bank bought SDRs from the government in exchange for long-term, non-transferable treasury bills, allowing SDRs to be included in the budget and used to repay the IMF during Covid-19.

SDRs have great potential but current use is limited and mainly reactive for crisis response. Domestic regulations, especially central bank rules, restrict broader SDR use due to fears of illegal monetary financing. This limits SDRs beyond reserve accumulation.

Our research, including work for the G20, highlights these issues and proposes reforms. Allocations based on quotas favour advanced economies over vulnerable ones. We advocate reallocating SDRs using vulnerability and financial need indicators. We also seek to reform domestic rules to allow SDRs more proactive financial roles. I’m happy to discuss more in the Q&A.

Jon: I want to zoom out a bit, because as Andrés showed in a more specific way, SDRs have always been a contested issue in the international monetary system. Since their creation in the late 1960s, there’s been debate over how SDRs should be allocated and whether there should be a development focus built into the system. Instead, we ended up with an imperfect system where SDR allocations remain linked to IMF quota shares. This link is very problematic, as the recent 16th General Review of Quotas gave an equi-proportional 50 per cent increase, deepening existing inequities by favouring the US, Japan, and Europe.

This connection between quotas and SDRs shows how the international financial architecture still needs reform. The lack of a development focus in SDRs has contributed to what José Antonio Ocampo calls an international monetary “non-system,” where low- and middle-income countries are forced to self-insure against volatile capital flows by holding non-SDR reserves.

Until 2009, SDRs were barely used, with only about $30 billion in circulation before two rapid allocations then. This coincided with changes in accounting rules, despite a 1978 amendment to the IMF Articles of Agreement calling for SDRs to be the principal reserve asset in the global monetary system – something still not realised. Fund staff often cite the Articles to limit SDR use, but this key provision is ignored.

In response to these longstanding issues, civil society put forward a broader SDR reform proposal last year, building on many past efforts. The proposal has three parts: first, an immediate new allocation to ease multiple crises (though politically difficult now); second, reforms to simplify and reduce the costs of using SDRs, including reverting to BPM5 accounting rules and reforming SDR interest rates; and third, renewed discussion on making SDR allocations more equitable, possibly moving away from quotas to annual or needs-based allocations linked to growth, climate, or vulnerability.

Currently, SDR reform is part of the ongoing negotiations for the Fourth Financing for Development Conference this summer. The current draft retains language on developing a new SDR playbook and reviewing their role in the monetary system. This broader discussion shows how SDRs fit into the wider international financial architecture reform agenda, where there is strong consensus that change is needed.

If the language stays in the final document, it will provide a head-of-state level mandate to continue these discussions at the Fund. Whether or not the Fund joins us publicly in these discussions, this dialogue must continue because support for expanding SDRs, including calls for annual allocations, remains strong. I’ll leave it there but wanted to broaden the scope and highlight how this specific issue ties into bigger reform debates.

Niranjali: For many countries facing a complete wipeout of development assistance for several consecutive years, it becomes even more crucial to explore and promote alternative tools.

Questions & answers: 

Representative from Money Justice Collaborative: Could you explain more slowly the accounting transactions between the central bank and treasury in Argentina, and why they had to be done that way?

Maia: It’s complex due to accounting rules. SDRs were registered at the central bank, but the treasury needed to use them to pay the IMF. So, we did a transaction where the central bank “lent” SDRs to the treasury. The treasury gives a treasury bill to the central bank; this is internal. We had to include these resources in the budget as they become a resource to pay debt. This accounting transaction may differ by country, but since the treasury holds the debt, it had to be done this way.

Representative from Global Governance Forum: Are the funds allocated from high-income to low-income countries through the [Resilience and Sustainability] trust (RST) largely debt or grants? Is this a real transfer or future liabilities? Because if it’s debt, it defeats the purpose, right?

Maia: The RST was a great idea initially, with a big push to rechannel SDRs to countries for health and climate. But in practice, it hasn’t delivered. Access requires an upper credit tranche (UCT) program, meaning countries must have a full IMF programme already, which limits access. It had caps on resources and quotas, and wasn’t replenished as expected. Changing SDR distribution would need Articles of Agreement amendments, which is tough, but other policies might help without those changes.

Jon: Yes, the RST is concessional with a 10-year grace period, based on SDRs rechanneled into a liquidity facility. Rechanneled SDRs backstop this lending. Many climate-vulnerable countries, including the V20’s official stance, want the UCT programme eligibility requirement removed, as it’s a burdensome requirement, limiting access.

Andrés: RST is not grants or wealth transfer, just new loans, though longer-term. Regarding SDRs or other currencies invested in the trust, there’s an encashment rule requiring a third of deposits to be held in reserve, limiting re-lending. Rich countries are hesitant to invest SDRs in MDBs or other instruments due to the need to keep them liquid, and this is another instance of the SDR accounting rules causing confusion and limiting investment decisions.

Representative from ActionAid Senegal: Finance decisions affect communities and shouldn’t be confined to ministries alone. Civil society often lacks understanding of SDRs and their potential. Could we debunk the SDR narrative and develop toolkits or manuals for easier understanding, enabling civil society to track and hold governments accountable?

Maia: There’s an IMF SDR tracker with maps showing usage, but it remains niche. Data availability needs better translation for public impact. Institutions and civil society groups have reports, especially in Latin America, but awareness is low. This meeting and more research and activism are crucial, especially now, as new allocations may be hard given geopolitical challenges. I’m open to meetings to make this less technical.

Andrés: The mechanics and accounting of SDRs were simpler when created (like “paper gold”), but BPM6 and other rules have complicated our understanding – even central bankers struggle. We must simplify SDR accounting for wider use. Also, consider the SDR as a unit of account, used by IMF, BIS, and some UN agencies. Making it more mainstream in international accounting would help, since currently, UN accounting is mostly in USD, despite IMF mandates for SDR use.

Representative from Boston University Global Development Policy Center: How can accounting changes be made? What discretion do countries have, and what IMF approval processes shape Balance of Payments Manual changes?

Andrés: Central bankers worldwide often call IMF reps confused about SDRs. The usual IMF advice is to keep SDRs parked in reserves to avoid risks and high interest costs, which acts as a deterrent. Activism, including a Latindadd handbook on fiscal use of SDRs, helped push the IMF to issue 2022 guidance on recording SDRs’ fiscal use. The main limit is ignorance—even specialised bankers are unaware. Regarding methodological changes, the IMF’s Balance of Payments Statistics Committee, with ECB, Fed, central banks, and UN reps, meets annually to discuss such issues. The Reserves Technical Task Team handles SDR accounting. The last major discussion was 2006; the next input deadline for BPM7 is September this year. We should urge them to treat SDRs as debt-free instruments again.

Representative from Oxfam US: With the US and Europeans opposing a new SDR issuance, and debates on who should get SDRs, what’s the political likelihood of new allocations? Could this debt crisis period be an opportunity to prepare for future openings?

Jon: Changing the IMF Articles of Agreement is a medium-term goal—unlikely now, though political shifts (like negotiations around a possible US quota increase approval in Congress) could change things. The recent SDR playbook discussion has opened up more systematic reform debates. The 2023 UN High-Level Advisory Board report influenced current talks at the UN. SDRs stem from Keynes’ International Clearing Union idea but took 25 years to develop; we don’t have that time now, so keeping the conversation alive is vital—especially as SDRs may soon be called upon again in a volatile global economy.

Niranjali: This is a critical moment with major uncertainty and volatility, affecting the US and the world. We don’t know what’s coming, but a no-cost tool like SDRs could become politically viable. We need to survive the present crisis but also keep a hopeful vision for the future. SDRs must remain on the table as a key part of the global toolkit, while we keep dreaming and working toward a better system.

Andrés: A famous economist said crisis responses depend on the ideas already “lying around.” It’s crucial to keep discussing SDRs, tweaks, and reforms because the next financial crisis could be huge. We must keep SDRs visible so they can be a ready option when the next big crisis hits.

Representative from SEATINI (Uganda): On negative balance of payments in developing countries, should this be considered when allocating SDRs? Many developing countries export raw materials, not finished goods, worsening their balances. Should GDP or reserves factor into quotas? The current quota system doesn’t serve developing countries well, especially amid a multilateral trade system that’s failing them and rising unilateralism.

Andrés: It comes down to “Make America Export Again.” If we can convince stakeholders that developing countries importing more (including capital goods from the US) will boost global trade and jobs worldwide (including in the US), SDRs can support that. Expanding trade is key.

Jon: The 16th General Review of Quotas was frustrating: no meaningful quota realignment has happened in 20 years despite a five-year update mandate. This is a core governance reform challenge civil society keeps pushing for. African and V20 member states remain underrepresented. The quota formula is mostly a reference for a political negotiation. Meaningful realignment looks unlikely now, though the IMF board is expected to propose possible pathways soon. Because quota reform is stuck, SDR reform is critical to tackle separately – including through decoupling allocations from quotas – as the current quota system isn’t fit for purpose, especially for African members.

Niranjali: The question of how to reform the SDR system and reallocate resources remains open. There’s some thinking on what to consider – like income level, vulnerability, resilience, or reserves – to guide needs-based allocations. But it also depends on the purpose: responding to climate shocks or other crises. It’s an area needing more thought, and I welcome ideas moving forward.

Representative from the Network Movement for Democracy and Human Rights: SDRs aren’t new loans, but citizens, not politicians, repay them. The IMF deals with governments, but we need it to consider citizens’ rights. CSOs need more capacity to hold governments accountable for these deals. Often, we don’t even know if those in power now will remain after elections, yet it’s the public that carries the burden. We need training to understand SDRs better. The IMF is slow on transparency and involving CSOs in negotiations. We’ll keep advocating, but it sometimes feels like the IMF wants countries to stay in debt.

Jon: This is why many civil society groups supported the 2021 SDR allocation – unlike traditional IMF programmes tied to conditionality and negotiated behind closed doors with little oversight, SDRs in a general allocation come without those strings. Countries can use them flexibly, based on national monetary rules. That’s a key difference. And yes, we need to improve outreach – translate materials on SDRs into more languages, provide more resources to CSOs.

Andrés: That’s why accounting matters – it shapes perceptions. SDRs aren’t typical IMF debt. The IMF itself excludes them from debt sustainability analyses because only a small interest is due, and there’s no repayment of principal. Misunderstanding them as debt scares policymakers off. But if we explain that SDRs are not debt- like the IMF used to call them for 50 years – more people might support them. That could help finance development without harsh conditions or austerity.

Representative from Global Governance Forum: I was wondering, for those of you who are very much engaged in the SDR debate, whether there is any discussion of also changing the SDR basket, because at the moment, we have a system where we have the five right, the sort of the economic equivalent of the Security Council of the UN. We have this privileged group of five countries that that are in the SDR basket. But you know, Canada, Australia, India, we have these large emerging economies, and we have the UK in the SDR basket, the UK is a very much diminished power post Brexit, with a shrinking economy over the medium to longer term and so on. So I’m just curious. The reason I ask this question is because there is an issue of equity and fairness, number one. Number two, if the SDR basket, let’s say, contained 10 currencies, it might create a greater sense of ownership among the membership, and then we might be able to get more countries to be engaged in this debate if they had a stake in it, right?

Andrés: In the 1980s there were 10 countries’ currencies in the SDR basket, but things were simplified after the creation of the euro and its addition to the basket. The main change in recent years was China’s currency, the renminbi, being added to the basket in 2016. Some people think that this was the result of a political negotiation which allowed the US to keep its veto at the IMF under the Obama presidency. So there is not a technical reason for which currencies are part of the basket, the reasons are geopolitical. Having more currencies as part of the basket would definitely improve the sense of shared ownership.