Spartacus Derivative Products
← Research & Perspectives
Paper 14

Bolstering sovereign debt sustainability: Integrating EMDE banks in the debt issuance and swap trading activities of MDBs

Nick K. Kolev Independent Consultant
Amil Aneja Independent Consultant
Development Finance Agenda · Volume 11 · September/October 2026 · © 2026 CIDEF
Summary
Since 2020, sovereign debt, measured either in nominal terms or as a percentage of economic output, has reached historical peaks. In some emerging market and developing economies (EMDEs), with official development assistance (ODA) decreasing, capital flight and sovereign default represent acute risks. Being countercyclical lenders, multilateral development banks (MDBs) can strengthen sovereign debt sustainability (SDS) by lowering their lending rates to borrowing member countries (BMCs).1 Through capital market development (CMD) and domestic resource mobilization (DRM), MDBs’ lending rates to BMCs can be lowered on market terms and without ODA dependence. This article describes an initiative – Spartacus Derivative Products LP (Spartacus) – that achieves this outcome by integrating EMDE banks in the debt issuance and swap trading activities of triple-A MDBs.2

MDB liabilities

In general, the capital market operations (CMOs) of development finance institutions (DFIs) and MDBs play important resource mobilization roles.3 By crowding private investors into DFI and MDB debt, they leverage the equity capital provided by shareholders.4 In addition, the CMOs of MDBs play vital SDS roles. Because SGOs are based on pass-through pricing, the lower the borrowing cost of an MDB, the lower the lending rates of the MDB to its BMCs.

As frequent issuers, MDBs strive to diversify their funding products and investor bases. They issue global bonds and private placements, and they rely on private and sovereign entities to purchase their debt.5 This is where declining ODA poses a material risk: sovereigns provide both equity and debt. Therefore, if it does not have the financial resources to support an MDB’s general capital increase (GCI), a sovereign is highly unlikely to purchase the MDB’s debt. The logic is straightforward. On the one hand, an MDB’s GCI is not associated with a major financial commitment because paid-in capital represents a small proportion – say, 7% – of a GCI. So, when subscribing to USD 100 million of new shares, a sovereign is required to pay only USD 7 million. On the other hand, purchasing MDB debt, which usually trades close to par value, is associated with a major financial commitment. To purchase USD 100 million of MDB debt, a sovereign is required to pay about USD 100 million.

It is common for an MDB’s annual borrowing program to eclipse its total paid-in capital, and MDBs’ borrowing programs have trended upward. As a result, some sovereigns hold substantial amounts of MDB debt. With SDS in mind, any reduction in the sovereign purchases of MDB debt must be adequately offset.

Global bonds and private placements

To be adequately offset, any reduction in sovereign purchases must neither prevent MDBs from completing their borrowing programs nor force MDBs into higher funding costs. One rational approach is to replace redundant global bonds with private placements. This approach is rational for two reasons: first, sovereigns purchase only global bonds, while private investors purchase both global bonds and private placements; and, second, global bonds, which are also known as benchmark issues, are more expensive than private placements, which are often structured notes.

In the case of global bonds, redundancy can be determined from their function as benchmark issues. Assuming that an on-the-run USD curve must be maintained, there is no compelling reason for an MDB to issue more than USD 1 billion at each of the 2-year, 5-year, 7-year, and 10-year points. Therefore, within an MDB’s borrowing program, any issuance of global bonds exceeding USD 4 billion is redundant. Replacing such issuance with private placements improves overall borrowing cost. Moreover, the improvement can be accompanied by superior funding product and investor base diversification, as well as by high-impact CMD and DRM.

Private placements are executed on a reverse inquiry basis. They are purchased by professional private investors, and they can be complex products. However, since MDBs are interested in after-swap pricing, so long as an MDB can price, book, and value a structured note and the attached swap, the primary variable of interest is after-swap cost. Consequently, the availability of swaps that can be attached to notes is a key constraint on the issuance of private placements.

Integrating EMDE banks

EMDE banks are not typically on MDBs’ swap counterparty lists. Usually, when it is on the swap counterparty list of an MDB, an EMDE bank has been granted an exception for narrow operations; for example, exceptions have been granted in the case of local currency (LCY) operations.6 Such exceptions do not maximize either CMD or DRM. In fact, they worsen MDBs’ counterparty risk exposures, and they raise questions about fair access.

Because they are the borrowers, MDBs are predominantly focused on managing the counterparty risks associated with swaps attached to bonds and notes.7 Thus, an elegant solution – a solution aligned with the interests of all parties – is to establish a highly rated swap house that allows EMDE banks to offer MDBs note/swap pairs.8

Alignment and elegance are intertwined when the swap house requires no ODA, operates on non-concessional terms, and utilizes market infrastructure.9 Owned by private institutions, offering products to private investors, and clearing its swaps through central counterparties (CCPs), the swap house offers liquidity, transparency, and volume.10

This swap house is Spartacus.

Spartacus

Spartacus is designed to carry zero net market risk and minimal counterparty risk. Hence, it can have high leverage and deliver abnormal returns for shareholders.11

The steps in the transaction process are the following:

  1. An EMDE bank offers a triple-A MDB a note/swap pair, with the swap between the MDB and Spartacus converting the MDB’s exposure into, say, USD floating rate;
  2. Simultaneously, the EMDE bank and Spartacus execute a mirror swap, ensuring that Spartacus carries zero net market risk;12 and
  3. Immediately, the swap between the EMDE bank and Spartacus is novated through a qualifying CCP, ensuring that Spartacus carries minimal counterparty risk.13

Every transaction follows this process. Spartacus’s only purpose is to integrate EMDE banks into the debt issuance and swap trading activities of triple-A MDBs. Its core mission is to lower MDBs’ lending rates to BMCs.

The integration facilitated by Spartacus has clear CMD and DRM benefits. Furthermore, Spartacus enables productive partnerships and reinvigorates faith in the multilateral system. Available to all EMDE banks on market terms, Spartacus ensures that every EMDE bank has the opportunity to lead-manage debt issues for MDBs.

This opportunity is dynamic in nature. CCPs offer central clearing for numerous products that are not mandated for central clearing. Also, they are interested in expanding their product offerings. Finally, they may be willing to combine existing CCP products to generate efficiency gains.

Conclusion

MDBs can strengthen SDS by lowering their lending rates to BMCs. Because MDBs’ lending rates depend on MDBs’ borrowing costs, MDBs seek ways of minimizing their borrowing costs. By integrating EMDE banks in the debt issuance and swap trading activities of triple-A MDBs, thereby helping MDBs prudently replace global bonds with less expensive private placements, Spartacus bolsters SDS while contributing to CMD and DRM. The development impact of Spartacus is immense because it requires no ODA, operates on non-concessional terms, utilizes market infrastructure, and works with all EMDE banks.

Footnotes
  1. An MDB is defined as a supranational institution that engages either exclusively in sovereign guaranteed operations (SGOs), such as Inter-American Development Bank, or both in SGOs and non-sovereign guaranteed operations (NSGOs), such as African Development Bank (AfDB).
  2. Basic information on Spartacus may be found here: https://spartacusderivatives.com/. Detailed information, such as responses to frequently asked questions and the financial model, may be obtained from the corresponding author: nick.kolev@spartacusderivatives.com.
  3. For example, see “Private-sector mobilization: recognizing the contributions of private-sector holders of MDB and DFI debt” (https://journals.co.za/doi/abs/10.10520/ejc-defa_v11_n2_a4). A DFI is defined as an institution, either national or supranational, that engages exclusively in NSGOs.
  4. Depending on the institution, equity capital may be composed of paid-in capital and callable capital. Both private and sovereign entities are investors in DFI and MDB debt. A DFI may have both private and sovereign shareholders. The main MDBs – the triple-A MDBs that are the subject of this article – have exclusively sovereign shareholders.
  5. The separation into global bonds and private placements is made mainly for the sake of tractability. For instance, MDBs issue Uridashi bonds, which are public secondary offerings to retail investors.
  6. When Mr. Kolev was at AfDB, AfDB’s Asset and Liability Management Committee granted such an exception to Standard Bank. Mr. Kolev worked with Standard Bank on the first supranational issue denominated in Nigerian Naira, the second supranational issue denominated in Ghanaian Cedi (the first one had been executed with Standard Chartered Bank), and the first supranational issue settled in South African Rand.
  7. Because an EMDE bank has to show an MDB a note/swap pair, execution risk can be considered insignificant. Even if execution risk is considered significant, MDBs must weigh this risk against their CMD and DRM mandates.
  8. DFIs and MDBs have created a swap house: The Currency Exchange Fund NV (TCX). The investors in TCX are almost exclusively DFIs, MDBs, and sovereigns. TCX’s mission is to promote DFI and MDB LCY lending by offering DFIs and MDBs swaps where markets do not offer them. Mr. Kolev was the inaugural Deputy Managing Director and Statutory Director of TCX Investment Management Company BV, and he was granted a fully delegated responsibility for the transaction process of TCX by TCX.
  9. When terms are non-concessional, there can be no market distortion. For example, see “Rational pricing in local currency non-sovereign lending” (https://journals.co.za/doi/10.10520/ejc-defa_v10_n3_a4).
  10. The contribution of volume to development impact cannot be overstated.
  11. By comparison, operating on the principle of diversification, TCX was designed to warehouse market risk. TCX’s counterparty risk exposure is higher than Spartacus’s. Spartacus’s target leverage is ten times TCX’s target leverage.
  12. Goldman Sachs Mitsui Marine Derivative Products LP (GSMMDP) operates on the principle of mirror transactions. GSMMDP, which has higher counterparty risk exposure than Spartacus, is on MDBs’ swap counterparty lists.
  13. Spartacus carries minimal counterparty risk because it has a triple-A MDB on one side and a qualifying CCP on the other side. The triple-A MDBs of interest to Spartacus are exempt from central clearing requirements.
Download the published PDF ↗