... little support for the legal and private litigation views, as the associated dummies are not positive as predicted but rather negative (and significant). This is not very surprising since in Table 2 we saw that auditors catch a mere 10.5 percent of the cases, while the litigation lawyers catch 3 percent.
I reproduce the relevant information from table 2 in pie chart form here.
What does work?
By contrast, we find strong support for the importance of the other three factors. As expected, detectors with monetary or career incentives are more likely to blow the whistle, as are detectors with better access to information.
I am struck by how little fraud is found by all types of government agencies. It suggests that the currently debated financial regulation overhaul is not likely to have a major effect on actual fraud.