In this article, we study firm-level determinants of corruption using a sample of 164,000 companies from 144 countries for the 2005–2020 period. We analyze two variables related to corruption: the perception of corruption as an obstacle to doing business using an ordinal logit model and actual bribe tax payments using a fractional logit model. Controlling for other factors, both sets of our empirical results show that the extent of corruption is related to the time spent dealing with regulations and inspections. We argue that firms which spend more time dealing with administrative procedures have a greater perception of corruption and are forced to make significantly higher bribe payments. Therefore, in a successful fight against corruption, it is essential to simplify administrative procedures by reducing their number and eliminating direct contacts between firms and officials.

在本文中，我们使用来自144个国家的2005-2020年期间的164,000家公司的样本研究了公司层面的腐败决定因素。我们分析与腐败相关的两个变量：使用序数logit模型将腐败感知为开展业务的障碍，以及使用分数logit模型将实际贿赂税支付。除其他因素外，我们的两组实证结果都表明，腐败的程度与处理法规和检查所花费的时间有关。我们认为，花费更多时间处理行政程序的公司对腐败的认知度更高，并且被迫支付高得多的贿赂金。因此，在反腐败的成功斗争中，