JONATHAN S. SEATON [*]
In this article the reform of academic tenure is examined in the United Kingdom. We test the hypothesis that reforming tenure may have reduced performance in the universities. The years following the 1988 Education Reform Act provide an interesting natural experiment, as the broad effect of the legislation was to soften (though not to remove) tenure in universities in the United Kingdom. It is concluded that the act has not adversely affected efficiency as some writers have predicted it would. (JEL I20, I28, K31)
In this article, recent university ratings in the UK Research Assessment Exercise (RAE) are examined with the aim of testing an efficiency theory of academic tenure. The results suggest that differences in performance exist between universities that could claim to have hard tenure before the 1988 Educational Reform Act and those that could only claim a softer form of tenure. Furthermore, the differences persist after the passing of the act, which suggests that tenure traditions have set up a culture that is not easily disrupted. The results accord with a widespread opinion in university circles that the act, which makes it technically possible to dismiss academic staff for financial reasons, did not really change things much. 
The passing of the 1988 Educational Reform Act gives us a good opportunity to test the efficiency-based account of academic tenure suggested by Carmichael (1988), who argues that it solves an incentive and information problem. According to Carmichael, the protection of late-career professors encourages efficient managerial decisions on their part. This testing is the first of its kind and is carried out using standard econometric methods. It should be of interest to those both inside and outside of academic life who may be concerned that increased managerial prerogatives within UK universities have had adverse unintended consequences on efficiency.
II. EFFICIENCY AND TENURE
A strong efficiency basis for tenure is given by Carmichael (1988), who argues that incumbent academics need an incentive to hire the best people if universities are to thrive. A university administration is dependent to a considerable degree on incumbent academics to judge the quality of junior staff. If academics thought they were vulnerable to replacement by better-skilled outsiders, they might hire low-skilled entrants to protect themselves. Therefore, tenure can be understood as creating an incentive for incumbent academics to reveal truthfully their judgments about the abilities of junior staff and to hire the best candidates available. This requires that the standard for firing incumbents be set exogenously and not in relation to new hires, but it does not imply that the university can never fire incumbents. Brown (1997) uses a similar approach to Carmichael's (1988) to explain tenure in terms of the need to motivate academics to act as monitors of university administrators in the setting of the evolu tion of boards of trustees in U.S. universities. 
An important assumption of Carmichael's (1988, 460) analysis is that university administrations do not know the value of alternative employment opportunities for weaker senior academics who, absent tenure, might be fired. Otherwise, they could simply devise a pension scheme paying the difference between the value of current and alternative employment opportunities, which should provide the same incentive. Another key assumption is that it takes time to discover the full extent of talent; otherwise, it could be cheaper to hire at random and fire weaker candidates rather than operate the revelation process described above.
According to Carmichael, the abolition or softening of tenure would have disastrous consequences for universities. In a tenureless academic world, incumbents would fill departments with low-quality academics to ensure that firing based on ability favored them. …