Edited by Susan Rose-Ackerman and Tina Søreide
Chapter 14: Evidence from the Firm: A New Approach to Understanding Corruption
Shawn Cole and Anh Tran
Shawn Cole and Anh Tran 1. Introduction Due to its clandestine nature, most of what we understand about corruption comes from survey evidence and self-reported perceptions of corruption. This limits both the range of questions that can be asked and the precision of answers that can be provided. This chapter proposes a new lens to understand corruption, using internal records collected from firms that pay bribes. We examine widespread corruption in three industries in an Asian developing country: procurement, pharmaceutical sales, and construction. Using data of real bribes, we provide new estimates of corruption and study its relationship with organizational ownership. Most current evidence of corruption comes from survey data, in which individuals and firms are asked about extra-legal payments made in the course of business and life.1 Survey data have a range of advantages: the harmonization of the phrasing of questions across surveys facilitates temporal and international comparisons; and surveyors have wide latitude in the type and nature of questions they ask (Kaufmann and Kraay, 2002). Surveys can also target representative samples: Svensson (2003), for example, found that a large fraction of Ugandan firms pay bribes, but that there was substantial variation in the size of bribe paid, even within industries. However, there are also well-known limitations to survey data, as recall is not perfect (Rose-Ackerman, 2006). Moreover, there is convincing evidence that individuals underreport the frequency and magnitude of their misdeeds (for example, Harrison and Hughes, 1997). Producers of surveybased data have taken these criticisms seriously: Kaufmann et...
You are not authenticated to view the full text of this chapter or article.