Apropos of an earlier post and discussion in class today about correlation and causality, here is an excerpt from an interview with Transparency International’s Huguette Labelle, where she answers questions about the apparent correlation between corruption levels and GDP, and corruption levels and levels of violent conflict:
The countries with the best scores in the CPI seem to be some of the world’s richest countries – is higher GDP the key to less corruption?
I think the difference between the countries at the top and the bottom is not so much due to their relative wealth or poverty, but to the development of their institutions. The top scorers have effective public sectors, with open contracting procedures, strong disclosure rules and access to information.
Labelle is implying here that the correlation between corruption and GDP is not causal; it is spurious (we’ll talk about spurious causation next class).
Many of the countries with the worst scores in the CPI are victims of violent conflict (Somalia, Myanmar, Iraq, Sudan and Afghanistan). What is the relationship between failed states and corruption?
In a crisis situation, the institutions of government are weakened, so corruption can more easily take hold and spread. It is not just individuals, but also institutions, that are responsible for maintaining integrity in a country. Many countries at the bottom of the CPI are failed states that are at the intersection of poverty, conflict and corruption.