GDP per Capita and Foreign Aid – Investment Frontier
We are endeavouring to provide a cursory glance at a very complicated and oft-researched subject; the impact of foreign aid on economic growth. Ordinarily, this would be certain folly, however we feel as though there are some insights to be gleaned.
The general conclusion of Economists is summed up well in the OECD paper, ‘Foreign aid, economic growth and efficiency development’ Ann Veiderpass and Per-Ake Anderson. They write, “In spite of numerous studies, there is little evidence of a significant positive effect of aid on the long-term growth of poor countries.” The paper itself focuses more on the idea that production may improve from foreign aid, and certainly there are improvements observable in some countries; notably Nigeria and China in the 60 countries they looked at. By and large though, there is limited evidence that even production was improved by foreign aid.
Economists have been debating the impact of foreign aid since at least the 1970s. Everybody from Papanek (1973) to Doucouliagos and Paldam (2009) have attempted to provide a unique perspective on the subject in a quest for proving that there is a link between foreign aid and economic growth; this quest has proved to be like the search for Eldorado.
So what are we bringing to the discussion? A simple analysis of how linked GDP per capita growth has been with foreign aid over time. The best case is where GDP per capita growth has been positive and foreign aid has been declining; the converse would be terrible. A positive correlation indicates mediocre effectiveness where both foreign aid and GDP per capita are increasing.
We found 16 countries that could be described as being successful, in terms of foreign aid and GDP per capita. That is a disappointing number, given we looked at the entire world’s data. However, it is not entirely unsurprising given the mountain of data to suggest that foreign aid is largely useless given how it’s delivered and more importantly who it is delivered to.
These 16 countries are quite interesting that this relationship is indicative of the presence of institutions and corruption, while present is not at a degree where it is debilitating to the country. It is important to note that our analysis is based on the past 25 years. We do not go further back due to the vast changes in the world since the 1960s (independence of most African countries) and the gradual demise of imperialism. Countries like Argentina look quite good since 1985, but a longer perspective would show how poorly the country has fared; going from one of the wealthiest countries to a middle-income developing country.
It is also important to note that 10 of the 16 countries are in the Americas. Clearly, something being done in the Americas is working. If we include Philippines as being within the American sphere of influence, that makes 11 of 16 countries. Only 2 countries in Africa, 1 in South Asia and 1 in Southeast Asia. It is also telling that no countries in the Middle East are shown despite being some of the largest recipients of foreign aid due to their inability to produce tangible benefits from the aid. The one exception is Tunisia which almost made this list.
Of the 16 countries, 6 are so small that they are unlikely to ever be looked at from an investment perspective. Of the remaining 10, 7 are on Investment Frontier’s dashboard (Philippines, Peru, Botswana, Indonesia, Argentina, Jamaica, and Colombia). The only ones we have overlooked are Mauritania, Angola, and Panama.
This is the way we approach analysis. We look at the simple, unadulterated data, and only then do we look for rationale that helps explain the data. We do not like looking at data initially after it has been sliced and diced by multi-factor models.
There are outliers in the data that ought to be ignored. For example, Afghanistan received over $5bn in aid annually, but this is largely focused on reconstructing a country that has been ravaged by three decades of war. The full data set is below. Please note that NA() errors are due to insufficient data.