Arbitrator Decision‐Making in the Transfer Market: an Empirical Analysis
| DOI | http://doi.org/10.1111/1467-9485.00053 |
| Author | Dennis Thomas,Alan Speight |
| Published date | 01 May 1997 |
| Date | 01 May 1997 |
II
NTRODUCTION
The analysis of alternative arbitration mechanisms has attracted considerable
theoretical and empirical interest in recent years. A particular motivation in this
respect has been the increasingly widespread use of arbitration in conflict
resolution, especially in the context of US public sector pay bargaining.1 From a
theoretical perspective, much attention has been paid to the intrinsic properties
of alternative arbitration procedures, and ‘conventional’ and ‘final-offer’
arbitration in particular. A key issue in this debate has concerned the relative
importance of ‘case facts’ and ‘final-offers’ in determining arbitrators’ awards,
and the potential for negotiation ‘chilling’ and the adoption of extreme final
bargaining positions prior to conventional arbitration in expectation of an award
that is some weighted average of the parties final-offers. However, whether
conventional arbitration procedures do in fact encourage arbitrators to indulge in
such compromise settlement behaviour remains an empirical question.
© Scottish Economic Society 1997 198
*University of Wales, Swansea
**University of Wales, Aberystwyth
Scottish Journal of Political Economy, Vol. 44, No. 2, May 1997
© Scottish Economic Society 1997. Published by Blackwell Publishers Ltd, 108 Cowley Road, Oxford OX4 1JF, UK and
350 Main Street, Malden, MA 02148, USA
ARBITRATOR DECISION-MAKING IN THE
TRANSFER MARKET: AN EMPIRICAL
ANALYSIS
Alan Speight* and Dennis Thomas**
ABSTRACT
This paper analyses data generated by the annual round of arbitrated
settlements of disputed transfer fees for end-of-contract players in the English
Football League. Using information on final-offers of the disputant clubs and
data on a variety of player and club characteristics we analyse arbitrated fees
and final-offers, and the relative importance of final-offers and case facts in
fashioning outcomes. Our analysis suggests that a large proportion of the
variation in settlements can be accounted for by certain case facts rather than a
mechanical compromise of final-offers, with the latter receiving a relatively low
weight. In general, our findings are consistent with the view that conventional
arbitration systems generate little useful information in terms of final-offers
from the disputant parties.
1For a review of the literature see Olson (1993).
Despite the wide range of contexts in which arbitration procedures are
employed, empirical studies to date have focused almost exclusively on labour-
management disputes, often involving experimental data on arbitrator decisions
in hypothetical cases.2 Whilst such studies afford the opportunity to design
hypothetical settings which ensure internal experimental validity, the external
validity of such studies inevitably remains in question.3 An alternative and
perhaps more obvious approach utilises field data on actual arbitration outcomes
in real settings.4 However, such studies are less prevalent since the final-offers
communicated by the disputing parties to the arbitrator are rarely made public,
and data relating to the circumstances of actual arbitrated cases may be difficult
to obtain. These difficulties are overcome here in the context of the annual round
of arbitrated settlements of disputed transfer fees for end-of-contract players in
the English Football League. This somewhat novel setting provides a more
extensive database for the analysis of conventional arbitration outcomes than is
typically available on several counts. First, the settlement decisions of the
Football League Appeals Committee (FLAC) provides a series of conventional
arbitration outcomes determined by a single body over a longer period than is
usually available. Second, the FLAC has recently made available to us details of
the final-offers put before the FLAC tribunal by the disputing parties at the point
of arbitration. Third, a wide range of data relating to the circumstances of
arbitrated transfer cases, including a variety of player and disputant club
characteristics, are readily available from existing published sources. Using this
data, this paper reports the results of an empirical examination of arbitrator
behaviour under conventional arbitration, with particular emphasis on the key
issue noted above, of the relative importance of case facts and final-offers in
settlement determination.
The paper is structured as follows. Section II provides a brief review of
alternative arbitration mechanisms and the debate concerning the relative
merits of conventional and final-offer arbitration, expanding on several of the
issues alluded to above. Section III outlines the empirical models of arbitrator
behaviour that are prevalent in the literature, and implemented here. Section
IV describes the operation of the Football League transfer market and the
procedure of the FLAC in particular. Section V describes the data set
employed and reports the results of implementing the empirical models
introduced in Section III. Finally, Section VI offers some concluding
comments.
IIARBITRATION MECHANISMS AND ARBITRATOR BEHAVIOUR
Arbitration is a widely applied method of resolving disputes that fail to be
settled by negotiation. While arbitration mechanisms can vary substantially in
ARBITRATION AND THE TRANSFER MARKET 199
© Scottish Economic Society 1997
2See, for example, Bazerman and Farber (1985), Farber and Bazerman (1986),Bloom
(1986), Olson et al. (1992) and Ashenfelter et al. (1992).
3See Anderson (1981), Gordon et al. (1984) and Olson et al. (1992).
4See, for example, Ashenfelter and Bloom (1984), Faurot and McAllister (1992), Burgess
and Marburger (1993) and Currie (1994).
To continue reading
Request your trial