Three years. That's how long it's been since the DFI Working Group last publicly reported on blended concessional finance. The latest public data from the group, which consists of more than 20 multilateral development banks (MDBs) and development finance institutions (DFIs), still stops at 2021. And when the next report eventually lands, it will only cover data through 2023.
Yet, in 2026, donors are being asked to keep channelling scarce public money including aid into a system running on a four-year-old snapshot. That would be hard to defend in any spending area. In one built entirely on public subsidy, it's indefensible.
Let's be clear about what this is
Blended concessional finance is subsidy. It uses public or philanthropic money to improve the risk-return profile of investments, making deals work that DFIs and private investors would otherwise walk away from. Done well, it can be a powerful tool. But with aid budgets being cut and donors making increasingly hard choices, they should not be expected to keep funding a system that reveals so little, so late.
And this isn't just a transparency problem. Limited disclosure is actively undermining the effectiveness of blended concessional finance itself. Transparency is a precondition for accountability, coordination, and learning. Without it, the private capital mobilisation agenda becomes harder, not easier, to deliver.
The problems run deep
At the project level—with IFC a notable exception—basic questions routinely go unanswered. Was concessional finance used? Why? How much subsidy was involved? Were alternative structures considered? Was the concessionality truly the minimum necessary?
In research I co-authored, we examined nine case studies across renewable energy, financial institutions, and agri-value chains in sub-Saharan Africa. A consistent pattern emerged: blended concessional finance often appeared to help get deals over the line, but the rationale for the subsidy was rarely documented with any rigour. DFIs can usually demonstrate financial additionality in their own-account investment, but the specific case for why concessionality was needed? Too often, that simply isn't there.
That matters. Development impact and DFI participation are not, by themselves, enough to justify public subsidy. There must be evidence that the subsidy crowded in the private sector in ways that not only make the current transaction bankable but also create future investments viable. If the unique need for concessionality isn't documented as a distinct decision, there's no way to tell whether concessional resources are being used in a disciplined, catalytic way, or just as a convenient deal making tool.
At the system level, things aren't much better. The DFI Working Group report is still the main consolidated public source. But it's too aggregated and too out of date to support meaningful benchmarking, learning, or informed policy debate. Donors and policymakers are making 2026 decisions using a 2021 picture. That's not a minor reporting gap; it's a structural failure.
The consequences are real
Poor transparency fragments the market. DFIs operating in the same sectors apply different structures and different concessionality levels, with no shared guardrails. That creates the conditions for a race to the bottom, institutions competing by offering softer terms, without any systematic way to compare or challenge those choices.
It also distorts markets in ways that rarely get enough attention. Hidden subsidy can produce fake market signals. Tariffs, financing terms, and return profiles may look commercial when they're actually subsidy-dependent. That misleads governments, distorts competition, and can deter new entrants. Blended concessional finance is supposed to build markets, not quietly warp them.
And it makes it harder to taper concessionality over time. If institutions can't learn systematically from peers or compare outcomes across similar transactions, they have no basis for knowing when, or whether, subsidies can be withdrawn.
Better disclosure is clearly possible
IFC has shown it can be done. Since 2019, it has disclosed basic information on blended concessional finance in public project documents: planned use of concessional finance, instruments employed, financing amounts, rationale for subsidy, expected development impact, and estimated subsidy levels. It's not perfect; explanations can still be high-level or incomplete. But it's materially ahead of most peers, and it proves that greater transparency is feasible even in commercially sensitive transactions.
Every MDB and DFI using blended concessional finance should be meeting at least this baseline. That's not a high bar. It's the floor.
So here’s the ask
Donors shouldn't just encourage better disclosure. They should require it as a condition of access to concessional money.
No transparency, no subsidy.
If an MDB or DFI wants donor concessional resources, it should be required to disclose, at minimum, whether concessional finance was used, what instrument was used, the size of the subsidy, the rationale for concessionality, how minimum concessionality was assessed, the expected market effects, and what actually happened afterwards. None of that requires revealing genuinely market-sensitive information. But disclosure does need to be timely, systematic, and comparable across institutions. A shared project-level reporting framework, covering both ex ante and ex post data, is the obvious place to start, and it needn't be fully public from day one. The GEMs database shows how this can work in practice: it began as a shared internal dataset among MDBs and DFIs, building trust and comparability over time before moving towards broader disclosure. A similar trajectory for blended concessional finance is both realistic and overdue.
This isn't an argument against blended finance. It's the opposite. If blended concessional finance is going to play a bigger role in mobilising private capital at scale, the case for it has to rest on evidence, discipline, and public credibility. Right now, that foundation is too weak.
Three years and counting is already too long. Donors should stop treating transparency as a nice-to-have and start treating it as a funding requirement.
