~ Pergamon Electoral Studies, Vol. 15, No. 4, pp. 461-476, 1996 Copyright © 1996 Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0261-3794/96 $15.00+O.(R) S0261-3794(96)00036-4 Constituency Campaigning in Parliamentary Systems with Preferential Voting: Is there a Paradox? 1 SHAUN BOWLER Department of Political Science, University of California, Riverside, CA 92521, USA DAVID M FARRELL AND IAN MCALLISTER Department of Government, University of Manchester, Manchester M13 9PL, UK This paper sets out to show whether constituency campaign effects exist, particularly in the context of a preferential electoral system such as Aus- tralia's. Data from the 1993 Australian Election Study suggest that, despite the best efforts of the candidates, local campaigning has a very limited impact on the vote. The paper considers a range of possible explanations for this limited impact: displacement, MPs' perceptions of their role, sources of campaign information, the importance of party labels in Aus- tralia, party discipline. It concludes with an explanation for the apparent paradox of why candidates bother campaigning when it does not make a difference to their vote. Copyright © 1996 Published by Elsevier Science Ltd. Introduction The extensive and growing literature on constituency, or district, level campaign effects has endeavoured to establish whether or not campaign effects exist in any great measure (for recent reviews, see Bowler et al., 1992; Seyd and Whiteley, 1995). However, much less attention has been paid to the processes by which these effects should exist. The theoretical basis for the argument that constituency level campaign effects should be muted is generally undeveloped. This paper--focusing on the Aus- tralian case--considers the issue of whether or not campaign effects may be said to exist in constituency-based parliamentary systems. In particular, we develop more fully than in previous literature the theoretical arguments which deny the existence of campaign effects. The Australian case also provides a useful example of the mediating influences of a preferential electoral system on campaign effects. To date, the bulk of research on campaign effects has relied on US and British examples. With few exceptions, attention has not been paid to the issue of whether campaign effects are influenced