People of dubious character are more likely to enter public service

People who entered the public sector had plagiarism scores that were 15.6% above the average.

According to new figures, 3.7m young Chinese sat the annual civil-service exam in November, a record. Some may not have had pristine motives for doing so. In a new working paper, John Liu of the University of Hong Kong and others use plagiarism in master’s theses to measure the level of dishonesty. Those who plagiarised more, they find, were more likely to go on to work in the public sector. Once inside they were likely to be promoted more quickly, too.

Illustration: Olivier Heiligiers

The authors come to this finding through a massive volume of data work. They trawled through 6m dissertations from CNKI, a Chinese repository of academic articles, and checked them against public records of civil-service exam-takers, identifying 120,000 civil servants and their dissertations. They used another 400,000 dissertations from non-civil servants for comparison. Messrs Liu and others ran the essays through an algorithm, producing a “plagiarism score” for each that described overlap with previous work.

The results are intriguing. People who entered the public sector had plagiarism scores that were 15.6% above the average. Customs and tax officials were the worst, with 25% and 26% higher scores than their private-sector peers. Once inside the civil service, cheaters climbed up the ranks 9% faster in the first five years of their careers than their more honest peers. From 2009 onwards plagiarism-detection software was gradually rolled out at universities across China. That reduced crooked behaviour, but those with higher scores were still more likely to enter the public sector.

The link between future dishonest behaviour and aggressive borrowing from published sources in past academic work may seem tenuous. But an experiment run by the authors seems to demonstrate that this variable reflects an enduring character trait. They invited 443 people as part of a job application to roll dice, rewarding them based on the outcome. The rolls were, however, unmonitored. People who plagiarised in their dissertations, they found, also reported improbably high rolls.

The authors explored the real-world impact of these tendencies by examining a database of 140m court rulings. Comparing the records of plagiarising judges with those of academically honest peers, they found that the cheaters were 10% more likely to favour the government over citizens in lawsuits, 15% likelier to rule for state-owned enterprises over private firms and 12% likelier to side with bigger firms than smaller ones. These judges, it seems, may have been susceptible to lobbying. When trials were livestreamed over the internet, these effects disappeared.

Performance evaluation in the civil service is notoriously subjective. Fudging metrics or sucking up to superiors may come easily to those with fewer moral scruples, which helps with promotion. Yet that does not explain why plagiarisers were more likely to enter the civil service in the first place. Worryingly for China, it may be that these people self-select into the public sector, suggests Shaoda Wang, a co-author on the paper, perhaps because they expect to thrive there.

Subscribers can sign up to Drum Tower, our new weekly newsletter, to understand what the world makes of China—and what China makes of the world.