Report
Tinkering with a Broken Policy: The IMF’s 2024 Surcharge Reform

Report
Fact-based, data-driven research and analysis to advance democratic debate on vital issues shaping people’s lives.
Center for Economic and Policy Research
1611 Connecticut Ave. NW
Suite 400
Washington, DC 20009
Tel: 202-293-5380
Fax: 202-588-1356
https://cepr.net
On October 11, 2024, the International Monetary Fund (IMF) announced a package of reforms to the charges and fees that it levies on its loans. Among the policies that were amended were those related to the Fund’s “surcharges” — fees imposed on countries whose outstanding debt to the IMF exceeds a certain threshold. The reforms follow years of accumulating evidence of harms caused by these surcharges and growing global calls to discontinue the policy, culminating in the initiation of a formal policy review process in July 2024.
While the IMF touted the October reforms as “substantially lower[ing] the cost of IMF borrowing,” what exactly are the consequences of these changes? What justification does the Fund provide for the reforms to its surcharge policy? Most importantly, do they address widespread concerns regarding the costs and impacts of surcharges?
In the following report, we estimate the new burden of the Fund’s amended surcharge policy and discuss the findings of the IMF’s policy review. We find that while the overall application of surcharges has been reduced, thereby lowering borrowing costs for a number of highly indebted countries, the policy remains onerous, procyclical, and counterproductive, in particular for the Fund’s biggest borrowers. Maintaining surcharges may be expedient for wealthy countries wishing to ease their commitments to Fund resources, but the evidence put forth in the review does not yield a compelling case that surcharges should remain in place. Ultimately, the October review represents a missed opportunity to ratify the precedent of the 1990s, when a similar set of surcharges was eliminated altogether, and was later acknowledged to have created “perverse incentives” by Fund staff.
Key findings:
Since 1997, the IMF has levied fees, referred to as surcharges, on borrowers whose debt to the IMF under the main General Resources Account (GRA) exceeds a certain threshold, defined as a ratio to the country’s quota at the Fund. In addition to “level-based” surcharges levied on all debt in excess of that threshold, a “time-based” surcharge is levied on debt that exceeds that threshold for a determined length of time — 36 or 51 months, depending on the lending facility.
As these surcharges specifically punish countries that are already heavily in debt, they have been criticized — including by CEPR researchers — as being procyclical and counterproductive, adding to already onerous debt burdens, siphoning valuable resources away from crisis-stricken countries, and making it more difficult for heavily indebted countries to reduce their debts to sustainable levels. In the years following the COVID-19 pandemic, a broad array of actors — including civil society, members of the US Congress, leading economists, UN human rights experts, and heads of state — advocated for the IMF to review, and in many cases to eliminate, its surcharge policy.
The IMF has reviewed its surcharge policies in the past. In 2016, there was a general review of surcharges that resulted in a decision to raise the threshold for level-based surcharges and lower the rate of time-based surcharges. There are also precedents for the IMF eliminating surcharges and surcharge-like policies altogether — in 1974, 1981 and 1992 — for failing to fulfill their purpose and, as the IMF argued in 2000, creating “perverse incentives.”Likely in response to the growing tide of criticism of surcharges, which surged during the COVID-19 pandemic, the IMF launched a formal policy review process in July 2024 which encompassed surcharges and other charges applied to its GRA lending.1 On October 11, 2024, the IMF announced the results of this review as well as a package of policy reforms to be enacted on November 1, which included the following:
In addition to these reforms to the surcharge policy, the IMF announced changes to other aspects of its lending framework:
While the reduction in the lending margin will ease the debt payment burden for all GRA-borrowing countries, the effective reduction in the lending rate at the time that the change took effect — even for countries not paying surcharges — was from 444 to 404 basis points. The headline rate for countries paying both level- and time-based surcharges decreased from 744 to 679 basis points. Figure 1, below, illustrates the monthly average headline rate of IMF borrowing, with and without surcharges, through November 2024, when the reform was implemented.
Figure 1
In the following section, we present updated estimates of the surcharge payments to the IMF following the October reforms. We then discuss the implications of the IMF review and reform package, specifically regarding the surcharge policy.
Post-Reform Surcharge Estimates
While the reform decreased the number of surcharge-paying countries in November 2024 from 19 to 10, the IMF projects that 13 countries will be subject to surcharges in FY 2026, as shown in Figure 2 below.2
Figure 2
By raising the threshold for applying surcharges, the IMF provided at least temporary relief for some countries, such as Suriname, Kenya, and Sri Lanka. Because these countries’ debt burdens with the IMF are projected to increase above 300 percent of quota, however, they will likely be subject to surcharges once again. Four additional countries are also at risk of paying surcharges in the near future given credit outstanding-to-quota ratios that approach the 300 percent threshold, as shown in Figure 3.
Figure 3
For the vast majority of the surcharge policy’s existence, the IMF did not publish information regarding which countries were subject to surcharges, or how much they were paying. In fact, even some country governments were reportedly unaware that they were paying surcharges. Beginning in 2021, CEPR has regularly published independent estimates of surcharge payments in an attempt to shed light on the opaque policy.3 In 2023, with surcharges under increased scrutiny, the IMF began publishing historical and projected surcharge payments through its online Financial Data Query Tool. However, the IMF’s method of projecting surcharge payments systematically underestimates their true cost over the long term.
As a result, it is still necessary to provide independent estimates. We have used available data to estimate the continued burden of surcharges from 2025 to 2030 (see appendix for methodology). Our estimates indicate that over the next six years, the IMF will levy $5.2 billion in surcharges. This is approximately $4.7 billion less than our previous estimate for the same period, $9.9 billion. This difference reflects both the reduced fees and the savings from countries no longer projected to pay surcharges. Argentina alone is expected to pay more than half of this total — $3.1 billion, down from $4.9 billion. Several other countries are still deeply impacted. Ukraine, for example, will pay over $860 million, and Ecuador nearly $640 million. Table 1 provides new, six-year total estimates for each country expected to pay surcharges as of FY 2026, compared to our previous estimates before the reform.
Table 1
Table 2, below, illustrates combined savings in surcharge payments and regular interest charges for the top five surcharge-paying countries over the next six years. While Argentina and Ukraine will be relieved of an estimated $1.7 and $1.3 billion in surcharges, respectively, Ecuador and Pakistan will see more meager savings of approximately $46 and $63 million, respectively.
In 2021, Argentina requested the elimination of surcharges. The table below also illustrates a counterfactual — how much these countries would have saved by today had surcharges been eliminated by 2022. In total, the top five surcharge-paying countries paid more in the fees since 2022 — $5.2 billion — than they will save in both surcharges and regular interest payments over the next six years due to the reform, approximately $5 billion.
Table 2The IMF notes that after these reforms it maintains “a considerable net income generation capacity,” with precautionary balances projected to remain comfortably above medium-term targets even in the event of adverse lending or investment shocks. According to the IMF’s projections, they will reach $35.5 billion by FY 2029 with the reform, as opposed to $51 billion. Indeed, with the IMF’s net income at record-breaking levels, the impact of the reforms on the health of the IMF’s balance sheet is marginal. In fact, as CEPR has previously estimated, the IMF’s income would have remained at a record-breaking high even under much more ambitious reforms.
While the reform package provided a welcome measure of relief, it fell well short of widespread demands for the complete discontinuation of surcharges, and leaves in place a policy that — as our estimates make clear — remains onerous to the countries affected. That the IMF elected to maintain the surcharge policy, and to stand by its procyclical logic, is particularly curious in light of the evidence presented as a result of the review process.
The IMF’s review corroborated a number of concerns regarding surcharge policy, including the fact that “the number of members paying surcharges has increased to a historical high” and that “borrowers are also paying surcharges for longer periods.” It further found:
In the current global monetary policy tightening cycle, the increase in the total cost of Fund borrowing for the largest borrowers has been more pronounced than for other GRA borrowers. While the increase in the total cost of Fund borrowing for surcharge payers since 2021 has been lower than the increase in the market cost of borrowing, the burden metrics for mean and median GRA borrowers have increased significantly since 2021.
The review emphasizes that Fund lending “is counter-cyclical in volume and in prices relative to market borrowing costs.” In other words, as might be expected of an institution whose goal is global financial stability, the Fund lends more in times of crisis, with borrowing costs that are “lower and more stable than market borrowing costs.” However, the review found that IMF lending is “not [counter-cyclical] in terms of its absolute cost. … Fund rates do increase when global interest rate [sic] rise, driven by the variable SDR interest rate. They also rise with the volume of borrowing and its duration, reflecting level-based and time-based surcharges, thus increase [sic] when members face particularly large and prolonged imbalances.”
In other words, surcharges are procyclical in nature, increasing the cost of borrowing for countries facing economic challenges, including those caused or exacerbated by exogenous shocks such as the COVID-19 pandemic. Given the considerable risks to the global economy in the coming decades, including from health crises, war, climate change, and economic fragmentation, surcharges are thus not only unjust for surcharge-paying countries facing these sorts of external shocks, but at odds with the objectives of maintaining macroeconomic stability and fostering sustainable development.
The Fund defends these procyclical and growing costs by noting that, even with surcharges, the cost of borrowing from the IMF remains lower than the cost of market financing. While this is true, the case is not made for why the market rate would be an appropriate bar by which to measure the cost of Fund lending. As a multilateral intergovernmental organization with a mission to promote financial stability and monetary cooperation, typically by acting as a lender of last resort to countries facing severe financial crises with limited market access, cost decisions should presumably be guided by their role in achieving these public goods, not in relation to the market’s profit-maximization incentive. Maintaining costs below market rates is a given, as otherwise the IMF would not have a role to play. It would be like comparing the low lending rates that the US Federal Reserve offered to corporations in 2008 to market rates at that time, taking into account the fact that many of these corporations had defaulted on loans or had bonds with junk status.
The most significant novel contribution of the surcharge review is an assessment of the policy’s effectiveness in disincentivizing large and prolonged use of Fund resources. Noting that “establishing conclusively whether the incentives provided by level- and time-based surcharges have discouraged excessive borrowing from the Fund and promoted early repayment is methodologically challenging because counterfactuals cannot be observed,” the IMF researchers reviewed the evidence of payment patterns surrounding the surcharge threshold. Ultimately, the researchers find that “countries facing moderate or short-lived financing needs are more responsive to level-based surcharges.” Countries that “have generally experienced manageable financing pressures” tend to maintain outstanding credit levels that hover just below the surcharge threshold, while countries that have borrowed moderately above the threshold during periods of need tend to reduce their credit to just below the threshold when financial pressures are mitigated. This is significant, as the IMF had not previously provided evidence of the effectiveness of the policy.
However, critically, the review also finds that for
countries experiencing high and protracted financing pressures … there is little evidence to suggest that the level-based threshold has influenced access decisions and for many of them Fund credit has remained high for a considerable time. These countries are currently having very high balance of payments needs and have no or only very limited access to alternative sources of financing.
This category of countries is responsible for over 90 percent of total surcharge payments, both before and after the reforms, as illustrated in Figure 4.
Figure 4
In other words, the countries that bear over 90 percent of the surcharge burden are the very countries for which the IMF admits that surcharges do not work. It is difficult to reconcile this finding with the decision to maintain the policy. Indeed, the top five surcharge-paying countries, Argentina, Ukraine, Ecuador, Pakistan, and Egypt, have repeatedly refinanced their IMF debt, and owed increasingly more in surcharges each year from 2014 through 2023.4
With regard to time-based surcharges specifically, the evidence is even more mixed: “The incentive mechanism of time-based surcharges appears to have worked for pre-2016 review cases related to the GFC [Global Financial Crisis] and for some post-review cases, while the evidence on effectiveness is less clear for more recent cases, mostly associated with the COVID-19 crisis.”
The cases specifically cited as successes, in which early repayments were made to avoid surcharges, are Hungary, Iceland, Ireland, Latvia, and Portugal before 2016, and Cyprus, Greece, Morocco, and St. Kitts and Nevis since. All but three of these — Iceland, Morocco, and St. Kitts and Nevis — are members of the European Union, and thereby have additional access to liquidity from the European Central Bank, and all but one — Morocco — are high-income countries. That this is a group of Fund borrowers largely of a different category to that of the vast majority of current borrowers is not discussed, though one footnote does note that the “Low global interest rate [sic] was conducive for large European borrowers to resolve balance of payment issues” and therefore to regain market access as an alternative to IMF lending — conditions that do not prevail today, particularly for countries that are not high-income.
In cases where surcharges may disincentivize taking on additional IMF lending, the review does not consider whether this is or is not necessarily beneficial for the long-term financial stability and development of the borrowing countries. In order to make the early repurchases necessary to bring outstanding credit below the surcharge threshold, countries may need to reduce spending in other areas, including potentially on necessary public goods such as education, health, climate action, or development.
For example, the IMF points to the fact that Morocco made an early repayment of $936 million (SDR 650 million) in January 2021 as its only example of the successful incentivizing effects of time-based surcharges among low- or middle-income countries. That very month and the following, the country suffered serious flooding which damaged infrastructure, destroyed homes, and killed dozens. The January 2021 early repurchase alone represents nearly two-thirds of the budget ($1.5 billion) of the country’s entire 20-year National Flood Prevention Program. In other words, surcharges, even if effective as a disincentive to further borrowing, may simultaneously incentivize increased austerity, with social and economic consequences that are not considered.
Finally, as the IMF notes, quota thresholds, which are fixed nominally, have a tendency to “erode” in real terms over time. By the IMF’s calculation, to counteract this erosion and bring the surcharge threshold back in line with its real value at the time of the last review in 2016, the threshold would need to increase to 280 percent of quota. Thus, the decision to increase the threshold to 300 percent of quota, while nominally a notable change, in fact does not go significantly beyond a return to the real value of the threshold in 2016.
In order to understand the reform package, it is important to consider how decisions are made at the IMF. The charge and surcharge reform was determined by the Executive Board, which is composed of 25 executive directors,5 each representing a country or group of countries, as well as the managing director. Each member country of the IMF holds voting power proportionate to its quota at the Fund. These quotas are in turn determined by a formula that considers economic size, trade openness, and accumulation of reserves, as well as informal political negotiations. In short, economic power is a major determinant of formal voting power at the Fund. The United States has by far the greatest formal vote at the Fund, with 16.49 percent. As many key Fund decisions must be made with 85 percent of the total vote share (though not the charge and surcharge reforms, which required 70 percent), the United States effectively wields veto power over much of the Fund’s activities.
While policy decisions such as these latest charge and surcharge reforms are often presented as a purely technical matter, they are fundamentally shaped by this balance of power. Stakeholders at the IMF are consulted during the course of the review, formal and informal discussions take place between members, and the board ultimately agrees to the final decision. While this decision is framed as a “consensus,” it is not a consensus of equal parties.
In the particular case of surcharges, the IMF, as well as its largest shareholder, the United States, have faced considerable pressure to discontinue the policy. In the past few years, calls for the long-term suspension or elimination of surcharges have come from UN Secretary-General António Guterres; UN human rights experts; dozens of former heads of state and government; the G77, and the G24, which represent the interests of nearly every developing country; leading economists; the UN Global Crisis Response Group on Food, Energy, and Finance; numerous leaders of Global South countries, including Barbadian prime minister Mia Mottley and Brazilian president Lula da Silva; and hundreds of civil society organizations worldwide, including Oxfam International, ActionAid International, Partners In Health, and the International Trade Union Confederation.
In 2022, the US House of Representatives passed legislation supporting a suspension and review of the surcharge policy, and in the weeks prior to the IMF’s announcement of reforms, over a dozen members of Congress, led by Reps. Jesús “Chuy” García (D-IL) and Joyce Beatty (D-OH), sent a letter to Treasury Secretary Janet Yellen calling for the elimination of the policy.
However, the US Treasury Department has resisted demands to discontinue or substantially reform the policy. One possible reason can be found in a discussion held in March 2024, when the target for the IMF’s precautionary balances, which are financed in part through income from surcharges, was soon to be met. While some members saw this as an opportunity to reduce or eliminate surcharges, others proposed maintaining surcharges and diverting surcharge income to finance the IMF’s concessional lending for low-income borrowers, which takes place through the Poverty Reduction and Growth Trust (PRGT). The PRGT is typically funded through voluntary contributions from advanced economies, but in recent years the PRGT’s financing needs have grown beyond wealthy countries’ willingness to contribute. By maintaining surcharges, but diverting their income to the PRGT, wealthy countries could effectively compensate for their own unwillingness to support low-income countries by charging highly indebted developing countries instead. Indeed, in April 2024 Treasury Secretary Janet Yellen directly proposed using GRA income to fund the PRGT. This was despite the fact that other options for funding concessional lending exist, even absent additional contributions from high-income countries.
While the process to arrive at this particular set of reforms was likely complex, in assessing the outcomes of the review it is worth looking beyond the mere technical decision that is presented in official IMF documents and considering as well the balance of power, and the motivations of the Fund’s largest shareholders, that underlie the decision.
While the IMF’s recent review of the surcharge policy at least temporarily relieved several countries from continuing to pay these fees, the meager nature of the reform means that surcharges will remain an unjustified, procyclical burden, especially for the Fund’s most indebted countries. The number of countries subject to surcharges, 10 as of the reform,6 is projected to increase to 13 in FY 2026, signalling the continuation of a vicious cycle.
The next review of the surcharge policy is scheduled to take place in 2029. Having recognized that surcharges do not function as intended for the borrowers paying 90 percent of them, the IMF Executive Board should take action to end this harmful policy sooner, as it has done in the past with similarly counterproductive fees.
Revised Methodology
In previous estimates of surcharges, CEPR calculated the GRA credit outstanding-to-quota ratio for all IMF member countries to identify surcharge payers as those with a ratio above the threshold (previously 187.5 percent, now 300 percent). In this estimate, we calculate surcharges for the 13 countries that the IMF announced will pay the fees in FY 2026. CEPR considers countries with GRA-to-quota ratios below 300 percent but exceeding 250 or 200 percent as having a “high risk” or “moderate risk,” respectively, of paying surcharges in the future.
Previous CEPR estimates were based almost entirely on data provided in the IMF’s staff reports for each country. Because the most recent staff reports were published prior to the reform, however, many of the numbers, like interest charges, are no longer valid. We now estimate surcharge payments for each of the 13 surcharge-paying countries based on a combination of still-relevant information from staff reports and the IMF’s Financial Data Query Tool.
Since 2023, the IMF has published projections of surcharge payments via this tool. Our estimates of surcharges, however, differ from the IMF’s query tool projections in one important aspect. The IMF projects future surcharge payments based on existing GRA credit outstanding, but not prospective credit, meaning that future disbursements of existing loans are not taken into account. For this reason, the IMF projections often significantly underestimate the true size of surcharges.
We begin our estimate by retrieving both the existing and prospective GRA credit outstanding per year from each country’s latest IMF staff report. To calculate level-based surcharges, we apply a 200-basis-points charge to the portion of this total above the 300 percent threshold per year. Time-based surcharges are more difficult to estimate, as they are applied to either all, or a portion of, the GRA credit outstanding above the threshold, depending on the amount of credit that has not been repaid within the limit of three years (or 51 months for Extended Fund Facility programs).
To estimate the amount of the credit outstanding to which time-based surcharges apply after the reforms, we first use the IMF financial data query tool to retrieve pre-reform projected time-based surcharge payments 2025–2030. (At the time of writing in November 2024, the IMF financial query tool still projected surcharges based on pre-reform methodology; i.e., the projections are based on the previous 187.5 threshold and 100-basis-point charge.) By dividing the projected time-based surcharge payments by the 100-basis-point charge, we find the amount of credit outstanding to which time-based surcharges applied before the threshold change. If this figure is greater than the total credit outstanding above the new 300 percent threshold, we can assume that time-based surcharges now apply to the entirety of the credit outstanding above the new threshold. We then calculate the time-based surcharges themselves using the new 75-basis-point charge.
If, however, the figure is less than the total credit outstanding above the new 300 percent threshold, we use the figure itself as the basis for the calculation of the new 75-basis-point charge. Because these estimates are based on the IMF financial data query tool’s projections, they do not include prospective credit, and should thus be considered an underestimate.
In one case, we assume that a country paying level-based surcharges will also begin paying time-based surcharges after their EFF agreement reaches more than four years of repayment, though this isn’t reflected in the IMF’s query data.
Given the assumptions noted above in estimating time-based surcharges, incomplete and non-standardized IMF records, and other factors, the figures calculated from this process should be understood as merely a rough estimate of real surcharge payments. However, we assert that they present a more accurate picture than the incomplete data contained in the IMF Financial Data Query Tool, which fails to include surcharges applied to prospective credit. The necessarily imprecise process by which we calculate our estimates would not be necessary if the Fund published all the data related to future surcharge payments.