Charter Cities vs Millennium Villages: What’s the Difference?

Noah Smith recently asked me on Twitter to compare charter cities to Millennium Villages.

Blog Posts


Noah Smith recently asked me on Twitter to compare charter cities to Millennium Villages. While there are some important similarities—namely, they’re both comprehensive attempts to create the conditions for sustained economic development and poverty alleviation— they differ in their ideas about how to alleviate poverty. This difference turns out to be significant.

Millennium Villages, first pioneered by Jeffrey Sachs, offer an integrated approach to fighting rural poverty. The idea: circumvent the poverty trap by simultaneously improving access to clean water, primary education, basic health care, and sanitation, plus other science-based interventions such as introducing improved seeds and fertilizer. The theory was that such a coordinated approach could create villages that would “end extreme poverty.”

Millennium Villages attracted tens of millions, maybe hundreds of millions (Wikipedia doesn’t give a total number), of donated dollars. George Soros alone donated $50 million to the effort. But Millennium Villages weren’t designed as a randomized control trial, which makes it difficult to judge the results.

Most estimates of the Millennium Villages efforts report at best, mixed results. Michael Clemens and Gabriel Demombynes found that “comparing trends at the MVP intervention sites in Kenya, Ghana, and Nigeria to trends in the surrounding areas yields much more modest estimates of the project’s effects than the before-versus-after comparisons published thus far by the MVP.” Michael Clemens tweeted that as a result of the article the Millennium Villages project threatened to sue him for defamation, which perhaps speaks to how those running Millennium Villages interpreted their results.

A Lancet article found that “substantial effects were seen in agriculture and health, in which some outcomes were roughly one SD better in the project villages than in the comparison villages. The project was estimated to have no significant impact on the consumption-based measures of poverty.” Another analysis found that for Millennium Villages in Ghana, “incomes did increase, but that this did not result in increased consumption. Instead, people appear to have viewed any income rise as a temporary phenomenon, with some saving in the form of liquid assets (e.g. chickens, guinea fowl, goats).”

In short, Millennium Villages did not meet their lofty goal of ending global poverty.

Charter cities however, could. Now, charter cities share similarities with Millennium Villages. As I previously noted, both use a  comprehensive approach to international development and poverty alleviation grounded in academic literature. However, they differ on two margins: in their understanding of the causes of poverty, and in their scaling mechanism.

The key intellectual difference between Millennium Villages and charter cities is their theory of poverty. Millennium Villages assume a poverty trap causes poverty. The poverty trap theory holds that a set of circumstances— low human capital, poor infrastructure, disease, etc— prevents capital accumulation, leading to continued poverty. Poverty traps can be escaped via a coordinated push which lifts people above subsistence, allowing them to save and create a growth cycle.

Proponents of charter cities on the other hand believe governance is the primary determinant of economic outcomes. Good governance leads to economic growth while poor governance leads to stagnation. Poor governance, meaning barriers to commerce, unpredictability of laws and regulations, and inadequate public goods provision, ensures that a population will remain impoverished. Improving governance then, is a necessary condition for economic success.

The second difference between Millennium Villages and charter cities is their scaling mechanism. Millennium Villages require substantial initial investment, usually in the form of donations. For example, the UK’s Department for International Development committed $18 million to fund five years of a new Ghanaian Millennium Village in 2012. The benefits, assuming there were any, from Millennium Villages are hard to capture, and because they require substantial amounts of donated capital, they’re even more difficult and inefficient to scale.

Charter cities also require substantial amounts of capital. Building infrastructure is not cheap. The difference is that the real estate developer captures some of the benefits of the charter city via the increase in land prices, making them able to attract investment, rather than donations. A successful charter city would generate large profits for the developer, incentivizing additional developers to build charter cities.

This is not to say that charter cities don’t require any charitable giving. I started a nonprofit because I believe it’s the most effective way to create charter cities. However, the required donation to start the first charter city is much less than $18 million and will steadily decrease over time with economies of scale. Charter cities are a mechanism for governance reform to help alleviate poverty that is designed to scale.

While there are certain similarities between Millennium Villages and charter cities, there are also important differences. Charter cities focus identify a different root cause of poverty, governance, and are built to rapidly scale. These differences are why I’m more optimistic about charter cities as a tool to help alleviate global poverty.