The MCC Wants to Work in Richer Countries. It’s a Bad Idea.

The Millennium Challenge Corporation, a US development agency with a budget of a little less than $1 billion a year, is facing increasing difficulties in spending those resources in the world’s 81 poorest countries. As a result, it is supporting a bill in Congress that authorizes the Corporation to expand country eligibility from the current requirement of low- or lower-middle-income countries to the world’s 125 poorest countries, which would include economies like Mexico and Malaysia. There’s some support for that move within CGD—including from my colleagues Nancy Lee, Erin Collinson, and Jocilyn Estes. But I think there are better ways to ensure MCC has the maximum development impact in the countries that would benefit from its support the most.

The existing MCC income cutoff is based on the robust principle that the best buys in global development are concentrated in the world’s poorest countries. (Indeed, when the Corporation started, it only worked in low-income countries for that reason.) There is a declining marginal utility of a dollar to deliver outcomes including health, education, or even life satisfaction: The first few dollars spent on vaccines and antibiotics buy a much larger increase in life expectancy than the last dollars spent on organ transplants. Name your measure of deprivation and you will find that it is almost certainly higher in poorer countries. These countries have far larger shares of people living in extreme poverty, suffering malnutrition, dying prematurely, and lacking access to basic infrastructure like sanitation or services like schooling. But the MCC suggests it is facing increasing constraints on its ability to spend its budget in the 81 poorest economies where that money will go the furthest.

This isn’t a matter of those countries’ ability to absorb external resources. The World Bank alone spends $36 billion a year in IDA credits and grants to countries with per capita incomes under $1,255 a year. That’s approximately fifty times what the MCC spends each year, in a candidate pool smaller than the MCC’s. Assuming, like income status, the new 125-country cutoff would be defined by Atlas GNI per capita, the proposed change would increase MCC’s income-based eligibility pool from a set of countries with a combined GNI of $8 trillion to a combined GNI of $29 trillion, based on 2020 numbers. Annual MCC compact and threshold program spending of about $682 million will fall from 0.008 percent to 0.002 percent of potential recipient GNI.

When the Corporation was founded, many in US development circles hoped it would provide a model for assistance programs, with approaches that would be rapidly scaled up within and beyond the agency. Now, the MCC claims it can't manage to disburse 2 percent of America's overseas development assistance (ODA) spending across countries home to half of the world’s population—just 17 cents per person per year. So much for a scalable model.

The MCC is having trouble even just staying the same size with its existing pool of income-eligible countries not due to a lack of need or opportunities, but because of the considerable constraints that its current model places on where and how it can work. To be eligible for an MCC compact, a partner country must perform relatively well on MCC’s ‘scorecard’ of development indicators and be in the top half of perceived corruption and democracy rankings for its income group, and the MCC must be statutorily allowed to work in the country. Even when countries meet those criteria, compacts have traditionally been capped at a certain size (topping out at about $700 million). Relieving those constraints would be the far better course to ensuring that the MCC keeps on working in the countries where its support matters most.

A first approach would be to reform the scorecard system to increase the number of low- and lower-middle-income countries in which it can work. I’ve written about the perverse impact and empirical weaknesses of the ‘corruption hard hurdle’ elsewhere, but the broader point is that when the MCC was created there was seemingly reasonable evidence that aid worked better in countries that were viewed as being ‘better governed’ with ‘better policies.’ That evidence is much weaker today.

Related to that, more than half of the measures in the MCC scorecard are fuzzy and subjective, open to considerable methodological critique, which makes them ill-suited as either screens for ensuring MCC resources are well spent or as yardsticks to encourage progress in order to qualify. Again, despite the fact the World Bank doesn’t use a scorecard-based eligibility mechanism, project ratings suggest Bank investments perform as well in the poorer (IDA) countries as the richer (IBRD) ones. That’s true even though poorer countries score worse across the range of MCC scorecard indicators. When the evidence changes, an organization that prides itself on evidence-based practice should change too.

Perhaps it is simply impossible to imagine Congress supporting improvements to the scorecard system (although MCC has managed to drop the disgraced Doing Business indicators it previously relied on). But even then, the MCC could do far more for the world’s poorest people simply by providing more financing in the countries that pass its current screening process: compacts could be significantly larger and more frequently repeated. Tanzania has been the recipient of the MCC’s biggest compact so far—a one-off program worth $690 million. It is simply implausible to imagine that Tanzania’s ‘constraints to development’ could be significantly removed with a single set of investments worth about one percent of the country’s GNI. It is simply implausible that the MCC ran out of projects it thought worth funding at that level: World Bank support to Tanzania in 2022 alone was more than twice that compact amount. Net ODA from all donors to Tanzania in that same (single) year was $2.3 billion. The original designers of the MCC envisioned an average compact size of $750 million-$1 billion, two or three times the actual average size of compacts, in part so that MCC would become the largest donor in many of the countries it operated. We are a long way from that vision.

And it is worth noting that if MCC compacts are too small to have impact in the world’s poorest countries, they’ll be drops in a bucket in the richer countries it wants to expand into. That’s another reason why larger, repeated compacts in poorer countries are far better than spending money at the top end of the new proposed eligibility threshold in countries that are fifteen times richer per capita than the average in the world’s low-income countries. In 2005, CGD’s Sheila Herrling and Steve Radelet complained when MCC decided to include lower-middle-income countries alongside low-income economies, noting: “[t]he promise of the MCC was that it would be innovative, entrepreneurial and transformational,” but pursuing more small compacts in richer countries “runs the risk of making it just another development program, which would beg the question of why we should incur the operational costs of a separate agency.” That argument is even stronger when it comes to working in upper-middle-income countries.

If the MCC needs more bureaucratic capacity to do faster, bigger repeat compacts in poorer countries (including support for repeat threshold programs), Congress should provide for that. One method to speed up compact delivery while reducing bureaucratic overhead would be to go back to the original intent and focus far more on budget support and far less on project finance.

But perhaps political realities mean the choice is between expanding the MCC’s country pool or seeing a smaller Corporation and the US spending even less on ODA. Even if that is the case, the bill is something to tolerate, not celebrate. It will make the MCC’s limited spending less effective. The Corporation has some attractive features (like the country-owned management group overseeing delivery) that it would be great to see operating in more than two percent of US ODA, but the bill is based on the assumption that these approaches can’t be scaled when it comes to the poorer half of the planet. It smacks of an indictment if that’s the best that the Corporation and Congress can manage. Holding fast to the MCC’s least evidence-based principles while abandoning the strongest in its search for investments is a sad look.