This work begins with a simple statement with big implications: “Institutions act through people.” It follows that the discretionary behavior of those in power matters, and when those people, be they CEOs, judges, academic leaders, managers, politicians, and others, act dishonestly, condemnation is often swift, though some escape retribution. This is especially true in the public sector, where global surveys identify “dishonesty” as the main criticism of politicians and civil servants. In other words, probity, or the strict adherence to the highest principles, is demanded of public servants, if not wholly expected.
Probity is so prized that it is instilled in students from an early age. Education systems worldwide enforce stringent oversight and severe penalties for academic misconduct, where a single instance of plagiarism of academic work can result in expulsion. And these charges can haunt a plagiarist for years, as even celebrated professionals face public disgrace years later when past misconduct surfaces. But are these celebrated cases just the tip of a plagiarism iceberg? What is the prevalence of academic misconduct and, further, what are the long-run impacts of such misconduct?
To address these questions, which are largely unexamined in the literature, the authors use China, home to the world’s largest higher education system and state sector, as a case study. Such an examination is empirically daunting for two reasons: First, by definition, undetected misconduct is unobservable, and you need undetected misconduct to separate the effects of hidden dishonesty from the effects of getting caught. Second, even if you can measure dishonesty, dishonest and honest people tend to sort into different careers and roles, making comparisons difficult.
The authors address these challenges by running plagiarism-detection algorithms across over 500,000 publicly available graduate dissertations. The algorithms go beyond simple copy-paste detection to catch paraphrasing and partial matches, while filtering out legitimate quotations. Since dissertations are legally required to be public and compulsory for graduation, this method delivers a clean, large-scale, and objective measure of high-stakes academic misconduct. The authors validate the measure using original survey data, finding that dissertation plagiarism appears to capture an individual’s underlying propensity for dishonesty.
By further combing dissertation plagiarism records with subsequent administrative data, the authors’ analysis of the data reveals the following:
- Public officials in the sample are meaningfully more likely to have plagiarized than their private-sector counterparts, with 19% exceeding the standard 15% copied-text threshold commonly used by most Chinese universities to withhold degrees, compared to 14% in the private sector.
- Moreover, within the public sector, exceeding that threshold is associated with 10–15% faster promotion rates, even after controlling for seniority, experience, and education. Rather than being screened out, academically dishonest individuals advance more quickly through government ranks.
- Analyzing over 140 million court verdicts, the authors find that judges who plagiarized their dissertations deliver significantly more favorable rulings to well-connected litigants, exercise discretionary power more liberally, and have their decisions appealed more often.
- Importantly, these effects are largely neutralized when court proceedings are livestreamed to the public, suggesting that transparency can act as a meaningful behavioral check.
- Dishonesty is not contained to individuals. Junior judges mentored by academically dishonest seniors gradually adopt similar ruling patterns over time. Meanwhile, lawyers who plagiarized their own dissertations become differentially more effective when their cases are assigned to judges who also have plagiarism histories. This pattern may reflect either top-down favoritism by plagiarizing judges or bottom-up capture by plagiarizing lawyers.
- As a result, as misconduct becomes more prevalent among peers, it becomes harder for honest individuals to function and advance, potentially locking institutions into low-integrity norms.
- Finally, while plagiarism-detection tools significantly reduced plagiarism rates, they do not alter adverse political selection: even under stricter screening, students with higher plagiarism rates remain more likely than their classmates to enter public service.
- That said, judges from cohorts exposed to stricter screening issue marginally fewer preferential rulings, suggesting that stronger enforcement of academic honor codes can produce at least modest long-run improvements in institutional integrity.
Bottom line: This work implies that academic dishonesty is widespread, particularly among those serving in roles pertinent to public welfare. Without adequate screening at the point of entry and robust checks and balances on the job, dishonest officials impose costs on society, both through their own behavior and by shaping the norms and incentives of those around them. Technology-assisted enforcement of academic integrity and greater transparency in institutional processes offer partial remedies.