Academic publishing is often framed around unethical authors, predatory journals, paper mills and citation manipulation. This framing though leaves out one important actor – the university themselves.
Rather than asking why some researchers behave unethically, it is worth asking what behaviours we should expect when universities, and the wider system, rewards publication volume, speed of publication, citations and indexed outputs. This is not an accusation directed at any particular institution or individual, but a question about how the system has developed such that it has generated these behaviours.
How incentives shape behaviour
In most universities, research performance is assessed using a similar set of indicators. These typically include the number of publications produced, where journals are indexed, their ranking or quartile position, citation counts and evidence of sustained publication activity. Such indicators are embedded in annual performance reviews, promotion processes, grant applications and institutional benchmarking exercises. Indeed, it is what the wider community also measures through measures such as rankings.
These indicators are attractive as they provide an apparently objective way to evaluate research performance. Moreover, they are simple to understand and can be reduced to single figures, such as an h-index.
However, metrics do not only measure activity. They also signal what is valued within the system. Once those signals become known, individuals respond to them in rational ways, aligning their behaviour with what is most likely to be rewarded. Many of the patterns we observe in publishing are logical responses to institutional expectations.
Measurements and targets
It is said that when a measure becomes a target, it loses much of its effectiveness as a measure. In academic publishing, this transition was made long ago, and scholars now target the measures, whether this is to increase your h-index, publish in Q1 journals or publish more than x papers a year – the measures that universities value are targets for its scholars.
When publication counts are emphasised, researchers maximise output. When speed is valued, there is pressure to publish quickly, sometimes at the expense of depth or robustness. When journal indexing and rankings are prioritised, attention shifts towards where work is published rather than the substance of the work itself. Over time, these signals accumulate and shape a shared understanding of what success looks like in academic careers.
The result is not necessarily misconduct, but a reorientation of behaviour towards optimising measurable outcomes. Over time, these become the norm and are accepted as such by both individuals and institutions. So much so, that they want to see year in year improvement on the measures, and therefore the targets also increase.
The predictable consequences
When viewed through the lens of measures becoming targets, many of the current challenges in scholarly publishing are easier to understand. Paper mills, for example, exist because there is demand for publishable outputs. Authorship markets emerge because authorship carries measurable value in promotion and evaluation systems.
Similarly, journals that offer rapid review and acceptance processes attract submissions because they align with pressures to publish quickly. Citation manipulation becomes more prevalent when citations are treated as indicators of influence and success. These developments are not isolated phenomena but interconnected responses to the measures, and now the targets.
In many cases, the individuals involved are not acting irrationally or maliciously. They are responding to the environment in which they operate, making decisions that appear reasonable given the expectations placed upon them and keeping up with their peers.
The grey zone of acceptable behaviour
One of the most difficult aspects of this issue is that much of the relevant behaviour exists in a grey area between clearly acceptable practice and outright misconduct. While certain actions, such as data fabrication or fake peer review, are widely condemned, many other practices are more ambiguous and often remain within formal rules.
Examples include dividing research into multiple smaller papers, prioritising quantity over depth, selecting journals based on speed of publication rather than rigor and engaging in strategic citation practices to increase visibility. These behaviours may not breach ethical guidelines, and in some cases, they are indirectly encouraged by evaluation systems that reward the outcomes they produce.
This creates a tension between formal standards of integrity and the informal incentives that shape everyday academic behaviour. It also makes it more difficult to draw clear boundaries between acceptable optimisation and problematic practice.
Institutional pressures and feedback loops
It is also important to recognise that institutions themselves operate under significant external pressures. Universities are assessed through rankings, funding frameworks and national evaluation systems, many of which rely on the same metrics used at the individual level but consolidated to reflect the quality of the institution.
As a result, institutions themselves should not be ignored when looking at incentives. They are participants, responding to incentives in much the same way as individual researchers. This creates a feedback loop in which institutional priorities shape researcher behaviour, which in turn reinforces the metrics used to evaluate institutions.
No single actor controls this dynamic, but it is collectively sustained. Understanding this feedback loop is essential if we are to move beyond treating symptoms and begin addressing underlying causes.
Why this issue is rarely addressed directly
Despite its importance, the role of incentives in shaping publishing behaviour is rarely discussed openly. One reason is that it challenges widely accepted approaches to evaluation and accountability. Metrics provide a sense of clarity and comparability and questioning them can be perceived as undermining established systems.
There is also an element of discomfort. If incentives contribute to problematic behaviours, then responsibility is distributed across the system rather than concentrated in a small number of individuals. This makes the issue more complex and less amenable to simple solutions.
Finally, there is the practical challenge of alternatives. While metrics have limitations, they offer a relatively straightforward way to assess performance at scale. Replacing or supplementing them with more nuanced approaches requires time, resources and careful design. And, changing the system, which is currently operating, is difficult.
Shifting the conversation
Addressing these challenges does not necessarily mean abandoning metrics altogether. However, it does suggest the need to rebalance how they are used. Greater emphasis could be placed on the quality and contribution of a smaller number of outputs, rather than the total volume produced. Similarly, increased attention could be given to transparency in research processes, reproducibility and the robustness of findings.
There is also an opportunity to recognise a broader range of contributions, including data sharing, methodological innovation and forms of impact that are not solely captured by citation metrics.
The key challenge is to design evaluation systems that remain practical while reducing incentives for behaviour that undermines research integrity.
The uncomfortable conclusion
Many of the issues currently observed in scholarly publishing are not anomalies. They are, at least in part, predictable outcomes of the way success is defined and measured within the system. While individual responsibility remains important, focusing solely on detection and enforcement is unlikely to be sufficient if underlying incentives remain unchanged.
This suggests that meaningful change will require attention not only to behaviour, but to the structures that shape it.
A question
If incentives drive behaviour, what would need to change to reduce demand for paper mills, authorship markets and questionable journals?
About the author
Graham Kendall is an experienced academic leader and researcher, with over 300 peer-reviewed publications and extensive experience across the United Kingdom and Asia. His work focuses on research systems, governance, artificial intelligence and the integrity of scholarly publishing.
He advises universities and research organisations on strategy, evaluation frameworks and research integrity. If you are reviewing how research performance is assessed in your institution, he welcomes the opportunity for a conversation.