Sunshine Works : Comment on "The Adverse Effects of Sunshine: A Field Experiment on Legislative Transparency in an Authoritarian Assembly"
Transparency -- sunshine -- is often touted as a core element of the governance agenda, and one that is most important in environments with low transparency to begin with. In a provocative paper published in the American Political Science Review, E...
| Main Author: | |
|---|---|
| Language: | English en_US |
| Published: |
World Bank, Washington, Dc
2013
|
| Subjects: | |
| Online Access: | http://documents.worldbank.org/curated/en/2013/09/18259607/sunshine-works-comment-adverse-effects-sunshine-field-experiment-legislative-transparency-authoritarian-assembly http://hdl.handle.net/10986/15831 |
| Summary: | Transparency -- sunshine -- is often
touted as a core element of the governance agenda, and one
that is most important in environments with low transparency
to begin with. In a provocative paper published in the
American Political Science Review, Edmund Malesky, Paul
Schuler, and Anh Tran present the results of a creative
experiment in which they provided an additional spotlight on
the activities of a random sample of delegates to
Vietnam's National Assembly. They report that the
effect of sunshine was negative, that delegates subject to
this treatment curtailed their speech, and that those who
spoke most critically were punished through the subsequent
election and promotion processes. The present paper argues
that Malesky, Schuler, and Tran's results, if
interpreted correctly, actually predict a net positive
effect of transparency. The differences in interpretation
stem primarily from three sources: the interpretation of
regression results for models with interaction terms, the
interpretation of the variable for Internet penetration, and
significant pre-treatment differences between treated and
control delegates. For the context in which more than 80
percent of delegates operate, Malesky, Schuler, and
Tran's results predict a positive but insignificant
effect of transparency. In addition, Internet penetration,
itself a measure of access to information, is positively
associated with critical speech. The paper draws lessons for
the design and interpretation of randomized experiments with
interaction effects. |
|---|