Corruption and Country Size : Evidence Using Firm-Level Survey Data
What sorts of conditions make some countries more prone to corruption than the others? This is an important question for understanding how corruption arises and how to combat it. The present paper attempts to answer this question by exploring the l...
| Main Authors: | , |
|---|---|
| Format: | Working Paper |
| Language: | English |
| Published: |
World Bank, Washington, DC
2019
|
| Subjects: | |
| Online Access: | http://documents.worldbank.org/curated/en/186421559050813620/Corruption-and-Country-Size-Evidence-Using-Firm-Level-Survey-Data http://hdl.handle.net/10986/31746 |
| Summary: | What sorts of conditions make some
countries more prone to corruption than the others? This is
an important question for understanding how corruption
arises and how to combat it. The present paper attempts to
answer this question by exploring the link between the size
of the country and corruption. Economic theory suggests
advantages and disadvantages of being a large country. Fixed
costs in monitoring and punishing corrupt politicians and
bureaucrats implies lower corruption in larger countries.
However, congestion or administrative costs may escalate
with country size. Further, greater diversity in the larger
countries implies that such countries may find it harder to
reach a consensus on growth-enhancing anti-corruption
reforms. Thus, the corruption and country size relationship
is an empirical issue. Using firm-level survey data for 135
countries, this paper finds that the level corruption
experienced by the firms is positively correlated with
country size. This holds for a measure of overall corruption
and petty corruption that arises in availing specific
government services. According to a conservative estimate,
moving from a country the size of Namibia (25th percentile
level in size) to a country the size of Morocco (75th
percentile level) is associated with an increase in the
level of overall corruption by 0.28 percentage point or
about 23 percent of its mean value. The results are robust
to several controls, alternative corruption measures, sample
alternations, and different country size measures. |
|---|