Does university management matter?

It's a commonly cited view that managing academics is like herding cats: difficult and ultimately pointless. But this view contrasts with growing evidence that good management increase productivity in both the private and public sectors.

We set out to examine whether the commonly held view is correct. Does management in universities matter? Are some universities better managed than others? And is better management associated with better performance? To do this, we took a tried-and-tested measure of management quality that has been applied to more than 10,000 organisations in manufacturing, retail, hospitals, schools, and social care for organisations from across the globe.

We used it to interview around 250 heads of departments in business, computer science, psychology and English departments entered for the 2008 Research Assessment Exercise in more than 100 UK universities. We complemented this with interviews with the heads of human resources departments in UK universities to get a measure of the quality of management at the central administration level. The survey covers management practices covering research and teaching processes, monitoring (performance measurement), targets and use of incentives (recruitment, retention, promotion and so on). The responses showed evidence of very good management as well as some less good practice.

One respondent, when asked how good performers were rewarded, replied that they were offered a congratulatory cup of coffee and some cake. Another, when asked how their university attracted staff, stated that their institution was simply the only place to work at and that everyone knew this. Another, when asked what their next post would be replied: "Death."

To those who have worked in academia for some time, these responses might strike a chord. However, there was good practice among the bad.

We examined the variation in management scores across and within universities and then examined the relationship between these scores and externally validated measures of research and teaching. For an overall ranking, we used the Complete University Guide. For research performance, we used the ranking achieved by each department in the government's Research Assessment Exercise of 2008. For student satisfaction, we used the National Student Survey scores.

The results show the following. First, in contrast to multi-plant manufacturing firms or even hospitals, university management is relatively decentralised: one department within a university can employ good management practices while another employs poor ones. But despite this, there are significant differences across universities in the quality of management practices.

Second, we defined four types of university: the most research-intensive universities (the Russell group), other 'old' (pre-1992) universities, former polytechnics and other 'new' universities. We found that the Russell Group scores highest, followed by the other old universities, the former polytechnics and the other new universities. While there are differences in resources across the types of universities, the results are not explained by this alone. Furthermore, the main driver of the difference in overall management score is in the areas of recruitment, retention and promotion. Performance in terms of targets and monitoring are much more similar.

Does any of this matter? The answer is yes. Departments that are better managed perform better, not only in terms of research, but also in terms of student satisfaction and the wider set of metrics as measured by the Complete Guide. Incentives matter for both research and teaching, but monitoring and targets are much less important in explaining performance. And the department level matters most for performance: the score of human resources departments at the same university adds nothing.

It's also not that there is one management style appropriate to the elite universities and another to universities that focus more on teaching and educating local students. Management matters in the same way at new universities as it does at old universities. In other words, good practice with respect to recruitment, retention and promotion improves rankings for universities that were former further education and higher education colleges just as much as it does for Russell Group universities.

On reflection this all makes sense: while universities deploy large bits of kit (science labs, and in some cases run hospitals), they are nevertheless people-dominated organisations. So getting it right with respect to staff matters.

Herding cats? Management and university performance was produced by John McCormick, Carol Propper and Sarah Smith – a working paper is downloadable here

Carol Propper is a professor at the Centre for Market and Public Organisation at the University of Bristol and Imperial College Business School

This content is brought to you by Guardian Professional. Looking for your next university role? Browse Guardian jobs for hundreds of the latest academic, administrative and research posts