For decades, the global financial community has treated sovereign debt crises like a broken thermometer: if we just improve the reading (transparency) and hold the technician accountable (through regulation and monitoring), the fever will break.
Yet, from the streets of Colombo to the manufacturing hubs of Dhaka, the fever is rising. As development finance dries up and emerging economies lean more heavily on market resources, the standard “good governance” playbook – pushed by the IMF and World Bank – is proving woefully inadequate to deal with sovereign debt.
Sovereign debt crises are often the product of a deliberately created ecosystem where domestic elites and external actors collude to generate unsustainable debt for private gain. To fix this, we must move beyond enhancing transparency and embrace a framework that understands the cold, hard realities of power, capabilities and interests.
The failure of vertical oversight
Traditional anti-corruption strategies rely on “vertical enforcement.” The theory is simple: a principal (the government or the public) uses information to hold an agent (a contractor, politician or bureaucrat) accountable.
But in many developing nations, this assumes a rule-following environment that doesn’t exist. Instead, actors are embedded in dense networks of informal negotiations where rule-breaking is the rational choice for those in power.
When a multi-billion-dollar infrastructure project is on the table, the incentives for kickbacks are so high that insiders can easily manufacture plausible justifications for inflated costs.
External monitors, no matter how transparent the data, often lack the technical capability or political muscle to challenge these sophisticated narratives.
A tale of two projects: Adani in South Asia
The recent experiences of Sri Lanka and Bangladesh offer a sobering look at how this collusion functions. In Sri Lanka, the Mannar wind power project – linked to the Adani Group – was flagged for significantly overpriced contracting.
While the levelized cost of electricity should have been roughly 5.58 US cents per kWh; the agreed price was 8.26 cents. This discrepancy represented a potential overpayment of over $600 million. Sri Lanka was “lucky” – a change in government and a public outcry forced a retreat.
Bangladesh was not as fortunate. The Godda coal-fired power plant, another Adani-funded venture, moved forward with a price roughly double the rate of electricity imported from the Indian grid.
With Bangladesh paying roughly $1 billion annually for the next 25 years, a 30% to 50% overpricing margin creates a massive, unnecessary foreign currency liability. These are no accidents.
They are the logical outcomes of systems where vertical oversight exists only on paper, while the underlying power dynamics favor collusion.
The case for ‘horizontal’ enforcement
If vertical oversight is failing, where do we turn? Our research suggests an alternative: a framework that identifies the Power-Capabilities-Interests of actors in the transaction ecosystem.
The PCI understands the rule-following/violation context as the net effect of formal and informal arrangements, both in terms of using relative power and influencing policy.
Policies will be feasible, implemented, sustainable and effective only if they build on and reinforce emerging behavioral changes and pockets of effectiveness arising from the organizational distribution of power and capabilities.
Designed in this way, policies will attract support from a critical mass of influential organizations, whose backing is driven by their own productive capabilities and interests.
Instead of trying to impose rules from the top down, we must design processes that are self-enforcing. This means empowering actors who have a direct, “selfish” interest in seeing the rules followed. We call this horizontal enforcement.
The key insight here is successful implementation is not achieved only by tightening enforcement, but by sometimes working with – rather than against – informality, understanding that one has to necessarily work with relationships beyond vertical principal–agent oversight, and aligning rule-following interests in such an ecosystem.
Horizontal enforcement works by creating checks within the policy flow. In more developed (high-income) economies, this often happens naturally because there are many rival firms and political actors who benefit from exposing a competitor’s corruption. In developing countries, we must build this ecosystem with targeted interventions.
Our research highlights two strategies that had notable impact in Sri Lanka and Bangladesh.
The first is splitting the rents: Instead of one massive mega-project that invites massive collusion, governments should shift toward a larger number of smaller-scale projects. This introduces more players into the market, increasing competition and making it harder for a single small group of insiders to control the narrative.
In Sri Lanka’s wind sector, as more investors entered the market with smaller technologies, the opportunity for corruption dropped alongside the prices.
The second is empowering productive rivals: By lowering the cost of capital for investors who participate in truly competitive bidding, we attract unconnected firms. These actors have a massive incentive to ensure their competitors are not getting sweetheart deals.
In Bangladesh, even modest competitive bidding reduced contracted prices by at least 25%.
From the bottom up
The goal is to change the ecosystem of project selection, debt contracting and project implementation from the bottom up. We cannot simply overlay a transparency sticker on a corrupt system. We must identify specific entry points – whether in project approval or implementation – where rival interests can be leveraged to act as a check on one another.
This approach is also relevant for other developing country contexts, including in particular, the African context, where countries face rising debt alongside declining competitiveness and limited productive employment.
By promoting domestically rooted investment and SME participation, African nations can create a class of stakeholders whose survival depends on the efficient delivery of infrastructure.
The era of relying on good intentions and better reporting to solve sovereign debt crises must end. We must stop pretending that information alone is power. Power is the ability to enforce a rule, and in the high-stakes world of sovereign debt, the only way to ensure the rules are followed is to make sure it is in the interest of the powerful to follow them.
Mushtaq Khan is a professor of economics at SOAS, University of London, and heads SOAS-ACE, the Anti-Corruption Evidence research consortium; Pallavi Roy is a professor in international economics at SOAS and the research director for SOAS-ACE; Ulrich Volz is a professor of economics and director of the Centre for Sustainable Finance at SOAS, University of London.
