Lurking Liabilities, Lego Contracts, and Low-Hanging Fruit
By Anna Gelpern, Dagmara Hanyz, and Katherine Shen, with research by Nona Tamale, Vanina Domizzi, and Vanessa Zou [1]
This Commentary and the accompanying STC Resource were last updated September 22, 2026.
Lurking liabilities
Public debt transparency is a rare point of multilateral policy consensus. Data commitments have featured in G20 summits, UN conferences, civil society campaigns and market initiatives since pre-pandemic days. Things have improved: low-income countries are getting higher transparency grades from the World Bank, bilateral creditors are starting to post restructuring terms, and multilateral processes look more intelligible with the publication of agreement templates. On the other hand, opaque contract designs are more popular than ever. Governments prize speed and secrecy especially when they come under liquidity pressure. Liquidity pressures are growing. Asymmetrically risky financial structures create an “invisible debt layer” faster than parliaments exercising oversight can make sense of them.
When debt terms are hidden, debts can disappear from sight, and debt numbers become guess work. Debt terms are easier to hide when debt transactions are made up of modular parts, negotiated separately—maybe months, years, even decades apart.
In a simple modular contract, standard terms sit in a virtual or literal file drawer and are periodically “incorporated by reference” in specific debt transactions, where they live between the lines. Standard terms act as debtor-creditor relationship constitutions and, like national constitutions, they rarely change. This leads some borrowers and accountability actors to assume that they are unimportant, immutable, or both, which makes their non-disclosure particularly pernicious. It also creates a rare opportunity for policy intervention.
Official creditors use the same modular approach to contract standardization as commercial creditors, but have public interest mandates and unique coordination tools at their disposal. Now is their chance to lead by example.
Each bilateral and multilateral creditor that uses standard terms and conditions (STCs)[2] in its public sector lending should publish them on its website in a consistent and accessible format. When making contracts, official creditors should ensure that their borrowers and relevant accountability actors, including parliaments and audit institutions, have timely access to the entire modular transaction bundle.
Publishing STCs and presenting all relevant documents together in one package would promote good debt management and debt governance practices, and make new borrowing and restructuring processes more efficient. As deliverables go, it is impactful and virtually costless—low-hanging fruit for the creditors, and an opportunity to practice the transparency they preach.
Lego contracts
Anyone who has borrowed money has heard the phrase “It’s standard.” It can soothe borrower nerves and make the jumble of words before them feel familiar, harmless, and in any case non-negotiable. This is wrong for at least three reasons.
First, “standard” does not mean identical or immutable. The very fact that terms are in a contract, not a treaty or statute, makes them presumptively negotiable. To know the scope for variation and the associated costs, borrowers need information on institutional and market practice from comparable contracts. The same lender used different confidentiality clauses with two different sovereign borrowers, two years apart. Different lenders use the same headings, such as “confidentiality,” “indebtedness,” or “negative pledge,”[3] for different contract mechanisms. Making this information accessible on demand in comparable format is an important motive for creating the #PublicDebtIsPublic Documentation Bank.
Second, “standard” does not mean trivial. All modern financial contracts are assembled from more-or-less standard parts, like Lego blocks. STCs incorporated by reference usually cover events of default and remedies, negative pledge and status undertakings, modification, confidentiality, and enforcement terms, among others. Each of these has played a critical role in a debt crisis.
Third, there are different ways of going about contract standardization, with different implications for the negotiating process and potentially different outcomes. For purposes of this Commentary and the #PublicDebtIsPublic STC Resource, we focus on two: contract templates and incorporation by reference, which may appear side by side in the same transaction.
A template usually looks like a contract with blanks for transaction-specific terms. The template is never the final agreement. It is a starting point for negotiations, and helps anchor the parties’ expectations of contract structure and content at a high level of generality. Loans traditionally require creditor due diligence and direct negotiations; loan contract templates therefore leave more room for customization. Sovereign bond investors usually expect more standardization, and rely on disclosure templates heavily influenced by capital market regulations. An analysis of annotated debt documents published on the #PublicDebtIsPublic platform at the start of 2026 supports the view that loans generally contain a wider variety of terms than bonds.[4]
Sovereign debt contract templates are usually produced by creditors, financial industry associations, or a mix. Most export credit agencies (ECAs), some official bilateral lenders, commercial banks, investment banks, trustees and others have “house forms” for different credit products, borrowers, and circumstances. Commercial bank loan templates generally follow the structure of templates produced by industry groups, such as the Loan Market Association (LMA) and the Loan Syndications and Trading Association (LSTA). The International Capital Market Association (ICMA) issues guidance and publishes handbooks for capital market instruments, some of which include model contract terms. These and other industry groups copyright their model documentation, and make it available to members and subscribers for a fee.
STCs incorporated by reference may look like full or partial contracts. In some cases, they are pre-printed. Their terms are then imported wholesale into another document that contains transaction particulars. When the parties choose to vary from the incorporated STCs, they spell out their variations in schedules, side letters, or financial transaction documents. Whereas templates are marked up, incorporation by reference keeps standard and negotiated terms separate. This can help focus negotiations and reduce inadvertent copy-paste errors, but can also be challenging to assemble and interpret.
Multilateral development bank loans, derivatives, and sale-and-repurchase transactions (repos) all use incorporation by reference, as do securities issued under a single trust or agency agreement.[5] Annex A includes illustrations from Jamaica’s loan and bond contracts, and Argentina’s repos.
In MDB loans, STCs commit borrowers to institutional policies, such as restrictions on secured borrowing, cancellation and acceleration events, and others. In derivatives and repos, core STCs are set out in a Master Agreement, published by the International Swaps and Derivatives Association (ISDA) and the Securities Industry and Financial Market Association (SIFMA), respectively. The Master includes events of default, early termination and netting obligations, governing law and enforcement, among others. An ISDA Master may be modified with negotiated schedules and protocols, and supplemented with a credit support annex, which governs the parties’ collateral and margin arrangements. Together, these documents are incorporated in individual transaction confirmations.[6] ISDA documentation is behind a paywall; the SIFMA Master Repurchase Agreement is publicly available on its website. MDB practices are uneven, as we elaborate below.
Some creditors revisit their STCs every few years; others hardly ever. Because new STCs incorporated by reference do not normally operate retroactively, earlier versions remain in effect for those contracts in which they are incorporated. For this reason, it is important to identify the version of STCs incorporated in a given debt contract, which can be difficult to find when it is not attached to the transaction terms. For example, at the time of this writing, the 2023 update of the World Bank’s 2018 General Conditions appeared to be incorporated in more than 200 agreements, but could not be found on the World Bank’s website.[7]
Terms Not Found
Official creditors’ STC publication record is surprisingly poor. Between March and September of 2026, #PublicDebtIsPublic researchers reviewed the practices of 186 agencies responsible for official lending in 70 countries. They found at least one version of STCs for one-third of 39 multilateral banks, for thirty four percent of 79 ECAs, and for just one bilateral development lender out of 93 responsible agencies they reviewed. Publication practices vary widely and for the most part unsystematically across countries and institutions.
During the same period, #PublicDebtIsPublic researchers located contract parts containing deal terms, but no corresponding STCs, on borrowers’ parliamentary websites. When STCs often account for more than half of the contract text, this raises debt governance concerns. What information do legislators see when they bless government borrowing? Do they have enough information to do their jobs? Do they know to ask?
Even when the STCs are published, most are practically inaccessible. The best creditor practice is a dedicated and intuitively labelled landing page in an intuitive location (eg, “Standard Terms and Conditions” under “What We Do” or “Products and Services”) that systematically lists and links to the STCs for each credit and insurance product offered and outstanding, as illustrated in Figure 1 below. Six multilateral development banks, 19 ECAs, and just one bilateral development lender follow this practice.
Figure 1: JICA General Terms and Conditions for ODA Loans
Without a dedicated landing page, using a keyword search (“standard terms,” “general terms,” “standard terms and conditions”) on a creditor’s website yields a version of STCs for six more multilateral development banks. For one multilateral institution, a general web search yields guarantee templates stored, but not displayed, on the creditor’s website.
STCs for some multilaterals, such as the Development Bank of Latin America and the Caribbean (CAF) and the Arab Bank for Economic Development in Africa (BADEA), could not be located on their respective websites. However, #PublicDebtIsPublic researchers found CAF and BADEA STCs on borrower websites, attached to individual transaction documents. Attaching STCs to the transaction terms in which they are incorporated at contract time and in subsequent disclosure is good contracting practice. Debt managers, parliaments, and the tax paying public in borrowing and lending countries should not have to waste time searching for matching parts of a modular contract in the wild, especially given the sad state of STC publication. Nonetheless, publishing transaction documents on borrower websites is not a substitute for publishing STCs on creditor websites. Both are forms of debt transparency, but they have different purposes and functions.
The primary purpose of contract publication by the borrowers is domestic accountability: the government is responsible to the public for the commitments it makes in its name. Standalone STCs represent creditor policies that may become part of borrower commitments. The primary purpose of STC publication by official creditors is policy transparency and accountability to stakeholders (borrowers, shareholders, and creditors, among others). Capacity building, norm diffusion, and contracting efficiency are important added benefits.
Low-hanging fruit
Public sector borrowers have an unprecedented range of conventional and exotic financing options, at the same time as many of them come under extreme liquidity pressures. Debt managers and accountability institutions in borrowing countries need real-time access to full-text contract terms to make high stakes decisions in the name of the public. Many more stakeholders, including the public in borrowing and lending countries and market participants benefit directly and indirectly from contract publication.
STCs published by official creditors are a valuable capacity building and accountability resource. Their publication can expose stakeholders to common instrument designs and contract terms, test claims that a given term is “standard,” and hold decision makers accountable. Access to STCs on demand, and before negotiations begin, can also help streamline new borrowing and restructuring negotiations.
Disclosure of standard terms does not require borrower consent. It need not implicate national security or commercial sensitivities: STCs have no transaction-specific financial terms or use of proceeds information. For official creditors, publishing every version of standard terms and templates in an accessible location on their websites, under intuitive labels, in a searchable format, is impactful and virtually costless. It is low-hanging fruit.
Common objections do not stand up to scrutiny:
“Standard terms are generic and therefore unimportant.” – To the contrary, STCs embody the core operating rules and principles of sovereign lending and borrowing, including disbursement and payment procedures, status and negative pledge, events of default, enforcement, and remedies. They are pervasive precisely because they are foundational, and need to be understood by a wide cross-section of stakeholders. Publishing STCs promotes norm diffusion and stakeholder ownership.
“Standard terms are standard and therefore non-negotiable.” – STCs are routinely supplemented and modified in small but potentially significant ways, using template markups, negotiated schedules, annexes, side letters, and transaction confirmations. Publishing STCs allows inexperienced and resource-strapped borrowers in particular to build debt management and negotiating capacity by studying contract practice, identifying variations, and evaluating the trade-offs.
“Official creditors with private sector development mandates follow private sector nondisclosure practice.” – Quite the opposite. Private sector borrowers from the so-called “private sector windows” of multilateral development banks routinely file full-text loan agreements with securities regulators and stock exchanges.[8] ECAs, which deal overwhelmingly with private sector borrowers, have a better STC publication record than multilateral institutions with private sector mandates. Some multilaterals published more information on their lending terms in the 1990s than they do today, when they have a unique opportunity to lead on debt transparency, at least with respect to their public sector lending.
“Official creditors publish handbooks and descriptions of their lending terms. Is it not enough?” No lawyer would advise their client based on a summary description in a world where one line of poorly drafted contract text could deplete a country’s foreign currency reserves. Official creditors should model informed decision making and responsible market practice, especially at a time when their borrowers are entering into risky complex transactions and parliaments approve them based on cursory summaries..
Official creditors have an opportunity to use their formidable coordination architecture – the G- fora, multilateral institutions, the Paris Club, and the Global Sovereign Debt Roundtable, among many – to promote transparency and good contracting practices. When publication is the default rule, government creditors can focus on coordinating among themselves to define exceptions for genuine commercial and state secrets.
There are welcome signs of movement: the Illustrative Template Memorandum of Understanding for Future Negotiations Under the Common Framework published by the G20 is a step toward more transparent restructurings, as is the publication of bilateral Memoranda of Understanding by the UK. They should set a precedent for broad-based routine disclosure of multilateral agreements and bilateral implementation.
The costs of opacity fall disproportionately on borrowers, but the entire system suffers. Without timely access to creditors’ STCs, sovereign borrowers cannot compare lender offers and make fully informed borrowing decisions; parliaments cannot scrutinize the terms their governments accept on their behalf; civil society and journalists cannot interrogate the structure of public debt portfolios; researchers cannot analyze trends across the development finance system at the depth its scale warrants. If the international development finance community is serious about debt transparency, publishing their own standard lending terms is the obvious place to start.
Annex A: Incorporation by Reference, Illustrated
Multilateral Development Bank Loan
Loan Agreement between the Government of Jamaica and the International Bank for Reconstruction and Development dated October 17, 2024, incorporating IBRD General Conditions for IBRD Financing: Investment Project Financing (Dated December 14, 2018, Last revised on July 15, 2023)
Standard Terms
Transaction Terms
Capital Market Debt
Indenture between the Government of Jamaica and Deutsche Bank Trust Company Americas dated July 28, 2015 and Global Note, Government of Jamaica Notes Due 2045 dated July 28, 2015.
Standard Terms
Transaction Terms
Sale and Repurchase Agreement (Repo)
Market Terms
Relationship Terms
Transaction Terms
ISDA Modular Transaction Suite
[1] We are grateful to Elsie Addo Awadzi, Lee C. Buchheit, Mitu Gulati, Sean Hagan, Nicole Kearse, Rosa M. Lastra, Marcelo Marzouka, Layna Mosley, Ugo Panizza, Stephen Park, Anahi Wiedenbrug, Jeromin Zettelmeyer, and the SDF researchers and partners who provided extensive comments on earlier versions of the STC Resource and Commentary.
[2] The #PublicDebtIsPublic STC Resource covers two kinds of Standard Terms: standard terms that the creditor regularly incorporates by reference or appends in their entirety to transaction-specific terms, and templates or forms of agreement that the creditor uses regularly in extending credit to the public sector. We refer to both as “STCs” in this Commentary.
[3] Different definitions of indebtedness may result in negative pledge terms extending to derivatives in certain bilateral and commercial bank contracts, but not in capital market or multilateral credit agreements.
[4] In theory, loan agreements are infinitely customizable because they are not actively traded. In practice, banks may have regulatory constraints and put a premium on exit options, which can limit the scope for cost-effective customization.
[5]See eg, “This Global Security is issued in respect of an issue of U.S.$ principal amount of 6.750% Notes due 2028 of Jamaica and is governed by (i) the Indenture dated as of July 28, 2015 (the “Indenture”) between Jamaica and Deutsche Bank Trust Company Americas, as trustee (the “Trustee”), the terms of which Indenture are incorporated herein.”
[6]The confirmation follows an industry template. In addition to the Master as modified and supplemented, it also incorporates industry-drafted definition booklets for different derivatives products (eg, interest rate swap, credit default swap, etc.).
[7] #PublicDebtIsPublic researchers searched for the document by name on the World Bank website, and found more than 200 documents that appear to incorporate it by reference. The STC document itself did not appear in the initial search results; however, it was initially identified in a Google search as part of a loan agreement published by the government of Argentina; it is excerpted in Annex A. The same version located, but apparently not indexed, on the World Bank website, appears lower down in the Google search results. Google search was used as a proxy for tools most likely to be available to everyday people seeking information on their governments’ lending and borrowing. Researchers checked results using Claude.
[8] Sovereign borrowers are subject to different, less onerous disclosure requirements.