Lancaster University Management School
Working Paper
2005/041
The impact of managerial quality on organizational
performance: Evidence from German soccer
Bernd Frick and Rob Simmons
The Department of Economics
Lancaster University Management School
Lancaster LA1 4YX
UK
© Bernd Frick and Rob Simmons
All rights reserved. Short sections of text, not to exceed
two paragraphs, may be quoted without explicit permission,
provided that full acknowledgement is given.
The LUMS Working Papers series can be accessed at http://www.lums.lancs.ac.uk/publications/
LUMS home page: http://www.lums.lancs.ac.uk/
THE IMPACT OF MANAGERIAL QUALITY
ON ORGANIZATIONAL PERFORMANCE:
EVIDENCE FROM GERMAN SOCCER
Bernd Frick
University of Witten/Herdecke
and
Robert Simmons*
Lancaster University Management School
* Corresponding author: Department of Economics, Lancaster University Management
School, Lancaster LA1 4YX UK, Phone 0044 1524 594234, Fax 0044 1524 594244, E-mail
r.simmons@lancaster.ac.uk
Abstract
Although a considerable literature exists on the determinants of managerial compensation,
much of it focussing on the role of incentives, there is much less known about the impact
of managerial remuneration and quality upon attainment of organizational goals. In this
paper we apply two distinct econometric methods to panel data on head coach quality and
team performance in Germany’s premier soccer league. We find that, given a particular
amount of spending on players relative to the rest of the league, a team that hires a better
quality head coach can expect to achieve improved performance.
Keywords: professional soccer, stochastic frontier analysis, managerial quality, or-
ganizational performance
JEL-Code: J44, L83, M50
2
1. Introduction
There is now an extensive literature on determination of managerial and CEO pay, much
of it driven by analysis of the classic principal-agent problem in modern firms. Prendergast
(1999) surveys a variety of research on the general issue of personnel and organizational
responses to alternative incentive mechanisms. In particular, Conyon and Murphy (2000)
examine the relative sensitivities of CEO pay to company performance in the US and UK.
Schwalbach and Graßhoff (1997) and Kraft and Niederprüm (1999a; 1999b) perform com-
parable analyses using data from German companies. Bertrand and Mullainathan (2001)
address the question of the extent to which relative performance evaluation determines
CEO salaries. The position of CEO salaries within company wage structures has also been
considered. For instance, Conyon et al. (2001) test hypotheses from tournament theory us-
ing British company data, showing that company wage structure is indeed highly convex
across top echelons of management. The relationship between different forms of corporate
governance and executive pay has received attention in a number of papers (e.g. Conyon et
al. (2001) and Hallock (2002)).
This research typically concerns the impact of company performance on managerial salary,
traced through theoretical impacts of incentive mechanisms on managerial effort. The re-
verse relationship between managerial quality and organizational performance has received
far less attention. This is not surprising. The data requirements needed to properly separate
out substitutability and complementarity amongst the host of inputs to firm outputs in
complex, modern organizations are usually too great to overcome and the risk of omitted
variable bias is bound to be high. Employer-based surveys have questions that are often
too broad to permit a focus on detailed technological and behavioural relationships be-
tween organizational inputs and outputs. Case studies using personnel records in particular
companies are helpful but results do not necessarily generalise.
As Kahn (2000) has persuasively argued, the sports industry is a useful sector within which
to test interesting hypotheses in the area of personnel economics. In professional team
sports, organizational goals and outcomes are much clearer than in most other sectors.
Teams usually wish to maximise sporting performance given available resources with which
to acquire playing and managerial talent. Increased sporting performance usually translates
3
into higher revenues and profits for team owners. In North American team sports in par-
ticular, there exists a plethora of individual performance and salary data often publicly
available on the internet. In Europe, where soccer is easily the dominant team sport, such
data have more restricted availability.
A general principle that can be applied to most sports leagues is that teams that pay higher
total salaries for their playing rosters relative to the league average will, on average, achieve
better performance. Performance can be measured by percentage of games won or points
achieved. In European soccer, owners and fans are likely to respond to position in the
league rankings since particular positions have further implications such as qualification to
higher level European competition. Empirical support for the general upward sloping rela-
tionship between relative team salaries and team performance has been found, using a
number of performance metrics and across a variety of sports leagues, in North America
and Europe, by Szymanski and Kuypers (1999), Szymanski (2000; 2003), Simmons and
Forrest (2004) and Kahane (forthcoming).
Even in the sports sector, studies showing the impact of managerial quality on organiza-
tional performance are sparse and limited in coverage. Some literature focuses on coaching
efficiency of sports teams as part of a more general treatment of team efficiency (e.g. Porter
and Scully (1982), Hadley et al. (2000), Dawson et al. (2000), Hautsch et al. (2001), Poulsen
(2000), Audas et al. (2002), Salomo and Teichmann (2002), Koning (2003) and Kahn
(2004)). This literature lacks direct evidence on coaching remuneration. Again, this is
largely due to lack of available data. In the case of English soccer, the league studied by
Dawson et al., the only wage data available for use as an input to team production was total
wage bill for the entire staff of the club. As the authors acknowledge, this total wage bill
measure conflates influences of player and coaching salary on team performance. Even the
pioneering study of managerial quality effects in Major League Baseball by Kahn (1993)
was hampered by only having a single season’s data for managerial salary (this also applies
to Kern and Süssmuth (2003)). Generally, coaching salaries in North American sports are
publicised far less than individual player salaries.
Our focus in this paper is on European soccer, specifically the top tier of German soccer
(henceforth Bundesliga 1) for which head coach salary data are publicly available. Within
4
