Working Paper
100

When Does Transparency Improve Institutional Performance? Evidence from 20,000 Aid Projects in 183 Countries

Date Published

Jul 14, 2020

Authors

Dan Honig, Ranjit Lall, Bradley C. Parks

Publisher

Citation

Honig, D., Lall, R., and Parks, B. (2020). When Does Transparency Improve Institutional Performance? Evidence from 20,000 Aid Projects in 183 Countries. AidData Working Paper #100. Williamsburg, VA: AidData at William & Mary.

Update: A revised version of this paper has been published in the American Journal of Political Science.


Announcement

Access to information (ATI) policies are often praised for enhancing transparency, accountability, and trust in public institutions, yet evidence that they lead to better institutional performance is mixed. We argue that a key impediment to the effective operation of ATI policies is the failure of public officials to comply with information requests that could expose poor performance. Unless accompanied by reliable mechanisms for preventing noncompliance, therefore, ATI policies are unlikely to improve performance. We test our argument through a difference-in-differences analysis of a new dataset on the performance of approximately 20,000 foreign aid projects financed by 12 donor agencies in 183 countries between 1956 and 2016—the largest dataset of its kind. We find that enforcement matters: ATI policies are only associated with better project outcomes when they include independent appeals processes for denied information requests. In addition, we recover evidence for several micro-level observable implications of our argument.

Bradley C. Parks

Bradley C. Parks

Executive Director

No items found.
No items found.