Researchers conduct studies, write papers, and share their findings with the wider community. Journals provide the mechanism to disseminate these papers. Peer review acts as a quality filter and citations build a cumulative body of knowledge by papers citing other papers which are relevant to the current study. This description describes how the system should operate, and it is important that it does operate in this way.
However, this is not the behaviour we now observe. If publishing were primarily about knowledge exchange, we would expect incentives to align closely with rigour, originality and long-term contribution. Instead, there is increasing emphasis on volume, speed and measurable outputs. This suggests that something more is at play, particularly in how research activity is evaluated and rewarded.
This leads to a simple but important question: what ae we actually rewarding?
The rise of a parallel economy
In recent years, several developments have moved from the margins of academic publishing into more visible territory. Paper mills produce manuscripts at scale, often with limited or questionable underlying research. Authorship markets allow individuals to obtain positions on papers without making a meaningful contribution – they may not even see the paper before it is published. Citation manipulation, whether through coordinated behaviour or editorial influence, can inflate the apparent impact of journals and individuals.
Paper mills respond to demand for publishable outputs. Authorship markets respond to demand for authorship positions. Citation manipulation responds to demand for citations and the metrics derived from them. When viewed together, these activities begin to resemble a parallel economy operating alongside the legitimate system of scholarly communication.
From papers to performance signals
To understand this shift, it is useful to reconsider what a research paper represents in today’s environment. Traditionally, a paper was a record of intellectual contribution, documenting a piece of work that added to collective knowledge. Its value was primarily intrinsic, grounded in the quality and originality of the research itself.
Increasingly, a paper functions as a signal, signalling productivity, the quality of the journal and how the citations demonstrate impact. Authorship position signals contribution and leadership. Journal indexing signals legitimacy, while metrics such as impact factors and similar indicators signal prestige.
These signals are embedded in evaluation systems used by universities, funders and governments. As a result, the paper becomes a vehicle for generating measurable indicators that can be compared, aggregated and used in decision-making. The signals associated with the paper become as important as, or sometimes more important than, the content itself.
Why signals dominate evaluation
One reason signals have become so influential is that they are easier to work with than the underlying substance of research. Assessing intellectual contribution is complex, time-consuming and often subjective. By contrast, counting publications or citations provides a relatively simple and standardised way to compare performance across individuals and institutions. Moreover, no discipline knowledge is required to count the measures, so anybody can make a judgement.
This practicality has led to the widespread use of metrics in promotion, hiring, funding decisions and institutional rankings. Publication counts, journal rankings and citation-based indicators are frequently used as proxies for quality and impact. While these measures can provide useful information, they also simplify a much more complex reality.
Once these signals are tied to career progression and institutional reputation, they acquire tangible value. They begin to function as a currency within the academic system, a currency that can be traded.
The logic of behaviour
When signals become the currency of evaluation, behaviour begins to adapt. If publication volume is rewarded, it is rational for individuals to seek ways to increase output. If journal prestige is rewarded, it is rational to target specific outlets, or to find alternative routes into them. If citations are rewarded, it is rational to maximise citation counts through any available means.
This does not imply that most researchers engage in misconduct. The majority continue to operate with integrity and produce high-quality work. However, the system creates pressures that shape behaviour at the margins. Under strong incentives, some actors will look for ways to acquire or enhance the signals that are being rewarded.
Paper mills reduce the cost and effort required to generate publishable outputs. Authorship markets allow individuals to acquire authorship signals without undertaking the underlying research. Citation manipulation increases the visibility of papers and journals within metric-based systems. Each of these can be understood as a mechanism for producing or amplifying performance signals.
Why the problem does not disappear
A common response to these issues has been to focus on detection and enforcement. Journals develop tools to identify problematic papers, databases refine methods to detect unusual citation patterns and institutions introduce policies aimed at discouraging misconduct.
These measures are important and can be effective in addressing specific instances of abuse. However, they do not fully resolve the problem. As long as the system continues to reward the same signals, demand for those signals will persist. Where demand exists, supply will emerge, often in new or more sophisticated forms.
Many current interventions focus on the supply side of the problem. They attempt to reduce the availability of manipulated outputs. What they do not address is the demand created by evaluation systems that prioritise measurable indicators of performance.
A system-level perspective
Viewing academic publishing through this lens requires a shift in perspective. The system is not only a mechanism for communicating knowledge; it is also a system for measuring performance. Both roles coexist, but the balance between them has evolved.
This perspective helps to explain why isolated interventions often have limited impact. Addressing one issue, such as paper mills, may simply redirect activity towards another area, such as authorship markets or citation practices. The system adapts because the underlying incentives remain in place.
It also highlights the importance of looking beyond individual behaviour. While cases of misconduct are often attributed to individuals, they are shaped by the broader environment in which those individuals operate. Understanding that environment is essential if more lasting solutions are to be found.
Implications for research evaluation
If the system is, in part, trading in performance signals, then improving research integrity cannot rely solely on better detection or stricter enforcement. It also requires careful consideration of how research is evaluated and rewarded.
This raises questions about the role of metrics in decision-making. Are current indicators being used helpful, or have they become targets? Do promotion criteria encourage meaningful contributions, or do they unintentionally prioritise quantity and visibility? Are institutions sufficiently aware of how their own frameworks shape behaviour?
These are complex questions, and there are no simple answers. However, they point towards the need for a more balanced approach to evaluation, one that recognises the limitations of signals while still making use of the information they provide.
The uncomfortable conclusion
There is a tendency to view paper mills, authorship markets and citation manipulation as external threats to the integrity of academic publishing. While this framing is understandable, it may overlook a deeper issue. These phenomena can also be seen as symptoms of how the system behaves under its current set of incentives. In economic terms, we have generated a set of unintended consequences.
If performance signals are the primary currency, then it is not surprising that strategies emerge to optimise those signals. The system may be functioning in a way that is consistent with how it has been designed, even if the outcomes are not what many would intend.
Recognising this does not diminish the seriousness of the problem. Rather, it shifts the focus from individual cases to the structure of the system itself. It suggests that meaningful change is likely to require adjustments to the incentives that drive behaviour, not just the mechanisms used to detect misconduct.
Your thoughts
If academic publishing is increasingly trading in signals of performance rather than purely in knowledge, what does that mean for how we evaluate research and researchers?
If incentives drive behaviour, what would need to change to shift the system back towards rewarding genuine intellectual contribution?
About the author
Graham Kendall is an experienced academic leader and researcher, with over 25 years in higher education and more than 300 peer-reviewed publications. He has held senior leadership roles including Vice-Chancellor, Provost and Deputy Vice-Chancellor. He is currently the Acting Vice-Chancellor at GlobalNxt University.