A Bosman ruling for science
2018; Springer Nature; Volume: 19; Issue: 11 Linguagem: Inglês
10.15252/embr.201847071
ISSN1469-3178
AutoresBrooke Morriswood, Oliver Hoeller,
Tópico(s)Sports Analytics and Performance
ResumoOpinion2 October 2018free access A Bosman ruling for science Brooke Morriswood [email protected] Department of Cell & Developmental Biology, University of Würzburg, Würzburg, Germany Search for more papers by this author Oliver Hoeller Freelance Science Illustrator [email protected] San Francisco, CA, USA Search for more papers by this author Brooke Morriswood [email protected] Department of Cell & Developmental Biology, University of Würzburg, Würzburg, Germany Search for more papers by this author Oliver Hoeller Freelance Science Illustrator [email protected] San Francisco, CA, USA Search for more papers by this author Author Information Brooke Morriswood1,‡ and Oliver Hoeller2,‡ 1Department of Cell & Developmental Biology, University of Würzburg, Würzburg, Germany 2San Francisco, CA, USA ‡These authors contributed equally to this work as first/second authors EMBO Rep (2018)19:e47071https://doi.org/10.15252/embr.201847071 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info With the World Cup over and the domestic football season now underway, the frenzied flurry of activity in the football transfer market is past, for now. All clubs (except Tottenham) have been getting their chequebooks out and doing deals, with players moving within countries, between leagues and across international borders. But what if scientists also commanded a transfer fee when they changed institution? It seems like a long time ago that Liverpool broke the Premiership transfer record by paying the then unheard-of sum of 8.5 million pounds for striker Stan Collymore, an amount that nowadays would barely acquire a squad player for a top-flight team. 1995, the year of Collymore's transfer, was arguably even more significant in footballing terms for something that happened in Belgium: the Bosman ruling. Marc Bosman, a journeyman player recently out of contract with RFC Liège, had lined up a move to French club Dunkerque. Dunkerque was however unable to meet the transfer fee demanded by RFC Liège. Consequently, RFC Liège refused to release him, while simultaneously reducing his wages because he was no longer a first-team player. Bosman sued for restraint of trade, arguing to the European Court of Justice that RFC Liège was preventing his free movement. He won his case. It was a ruling that transformed football. From then on, out-of-contract players became free agents who could move without the imposition of a transfer fee. Clubs suddenly had an incentive to either hold on to their players or sell them before they left for nothing. It is by no means an understatement to say the Bosman ruling revolutionised football and paved the way for the global business it has now become. The ongoing rise of transfer fees may not unequivocally be a good thing (Neymar's move to Paris Saint-Germain for 222 million euros has opened an eye-watering new chapter for those uncomfortable with the sums of money paid to grown men who kick a ball for a living), but it is indubitably the case that players currently hold the balance of power. Prior to the Bosman ruling, all power was with the clubs instead of the players; now the boot is on the other foot. That is a situation that scientists, particularly young scientists, can look on with envy—and one that we should perhaps try to emulate. For all the fine words about promoting the lot of the next generation, morale in the community remains low and long-term prospects dim. A key reason for that low morale is a lack of job security. The postdoc period has gone from being a 2-year stint abroad to gain more experience, to an open-ended limbo (purgatory for some) with no guarantee of a happy ending. The research environment has meanwhile become hypercompetitive to a toxic degree, meaning that doing good work is no longer a guarantee of professional success. And the Holy Grail of a permanent contract, a tenured position, is something that is now commonly not attained until a scientist reaches their mid-40s. Even for the most successful postdocs, landing a tenure-track position is elusive. It is not unusual that institutions operate a 5+3-year contract system for junior faculty with no possibility of tenure at the end—a setup that has the cynical benefit of harvesting the overheads from a scientist's first grant, but with no need to bet on their long-term success. If their early promise does not bear fruit or if they are simply unlucky, well, they will be out the door in 8 years and another crop of successful postdocs can be hired to replace them. It is all a far cry from the halcyon days of the mid-20th century, when the rapid expansion of the university sector meant that lecturers would sometimes not be much older than the undergraduates they were teaching, and permanent positions could be attained before the age of 30. Nowadays, young scientists are apt to find themselves on 2- to 4-year contracts, possibly dependent on raising some or even all of their own salary, and with no obligation for sympathetic treatment at the end: psychological stressors that are unlikely to make them produce their best or most creative work. But just imagine what it would be like if a transfer fee system was in operation. What if, when a scientist under contract moved from one institution to another, the hiring university/institute had to pay a transfer fee to the selling university/institute? The implications would be profound. First, it would incentivise universities to develop and invest in their junior faculty, thereby making sure that they were worth more if and when they did leave. The size of the transfer fee would be negotiated based on both institutions’ estimations of the faculty member's value—just as in football. While research performance would probably be the dominant consideration, it is not out of the question to imagine an institution also placing a high premium on someone with a reputation as an educator or communicator. Second, it would create a marketplace for faculty. Immediately there would be an incentive to offer longer contracts, maybe even early tenure, to rising stars. Any university that did not tie down the best of its younger faculty with long-term deals would get nothing when those same faculty members contracts’ expired and they just left. Out-of-contract free agents would mean no transfer fee for their former institution and therefore a loss of revenue. It is important to note that this would not mean that universities would necessarily need to start finding money for more permanent positions—though that would be no bad thing—but at the very least, they would be more likely to offer longer contracts. Plus, the increased revenue from transfer fees would facilitate more and longer contracts. The more successful a university became at developing young faculty, the more it would benefit. Currently, the only “customers” in the university sector (in the British and American systems, there is arguably no need for quotation marks) are students, so the emphasis for faculty is on teaching, meaning that finding additional time to conduct research can be a struggle. Introducing transfer fees would automatically turn other universities/institutes into customers and generate a second revenue stream when faculty were snapped up by other institutions. Therefore, ensuring that faculty maintained an active research programme alongside their teaching duties would be an essential way of ensuring sustained or enhanced value in the transfer market (Fig 1). Figure 1. A transfer market for scientistsThe introduction of a transfer fee for recruiting faculty could help to improve the career prospects of young scientists. Download figure Download PowerPoint It is already the case that powerhouse universities and lavishly funded research institutes have a strong pull for faculty members. Currently, the recruitment of productive faculty by such organisations tends to breed resentment from the institutions unable to compete with the packages on offer, and they become resigned to losing their best people. A transfer fee system would not change this flow, but it would mean that smaller universities that sold on faculty to the Harvards of this world would see a profit from it. Well-funded institutions would have to pay for the privilege of grabbing talent from other institutions, just as in football. In fact, it might even create a system where smaller universities are actively hoping that their faculty ultimately move on to larger, better-funded institutions, similar to feeder clubs in sports that nurture young talent. Hiring young faculty, developing them into group leaders with a strong pedigree in research, education and probity that makes them premier acquisitions for top-tier universities and then selling them on at a profit could become a lucrative business model. And gaining a reputation as a good incubator for emerging talent would help to attract more ambitious young scientists. In other words, almost everyone would gain: scientists and universities. Lastly, transfer fees would intrinsically encourage mobility of people within the system, something that everyone is generally keen on. A win-win. Are there risks? Of course: you may get institutes kicking out or discouraging people that do not look like good prospects, but that is pretty much what happens already. It would also inevitably encourage the arrival of superstar transfers in the Ronaldo, Mbappé and Neymar mould—but again, the current fixation on breakthrough science and the cult of the individual is arguably already here. It would thus make little difference to that echelon of scientists, but simply attach a money value to people's sense of self-worth. Importantly though, transfer fees would improve contract prospects for young scientists, who would undoubtedly account for the overwhelming majority of such deals, and that is surely a good thing. In funding circles, it has become a mantra to focus on people rather than projects (always so for the Howard Hughes Medical Institute, more recently for the Wellcome Trust too). But while people rather than projects have become funded, it has only indirectly turned scientists into assets based on their likely ability to attract funding. Transfer fees would make them a direct asset for their host institutions. Here is a call out to the big universities and institutes out there: the next time you recruit faculty, why not offer to pay a transfer fee? You might start something big. Brooke Morriswood is a junior group leader at the University of Würzburg; Oliver Hoeller is a freelance science illustrator based in the California Bay Area. Together they produce the science blog Total Internal Reflection (https://totalinternalreflectionblog.com). Previous ArticleNext Article Read MoreAbout the coverClose modalView large imageVolume 19,Issue 11,November 2018Caption: Using single‐cell transcriptomics and multicolour flow cytometry analyses, Sousa et al identified and characterized discrete microglial programs upon acute systemic inflammation. In LPS‐injected mice, microglia (IBA1 positive cells, in green; cell nuclei counterstained with Hoechst, in red) segregate into two distinct transcriptional reactive states. This study provides insight into microglia heterogeneity and establish a resource for the identification of specific phenotypes in CNS disorders with an immunological component. By Carole Sousa, Alessandro Michelucci and colleagues: Single-cell transcriptomics reveals distinct inflammation-induced microglia signatures. Volume 19Issue 111 November 2018In this issue FiguresRelatedDetailsLoading ...
Referência(s)