examining the role of valence factors across countries, and across time.
one of the major reasons for this lack of comparative work on valence issues is the
absence of measures of valence that can be utilised in cross-national studies. Given the
lack of readily available measures, it is not surprising that studies often focus on the state
of the economy or on incumbency, since measures for these two concepts are relatively
easy to obtain. To date, little research has focused on examining the electoral effects of
events such as political scandals, visible disagreements among party members, and
perceived incompetence.
As a result of these gaps in the literature, several interesting questions remain
unanswered. Can a cross-national measure of the relative impact of valence factors in
general elections be created? If so, will it confirm that events such as political scandals,
party divisions, and perceived incompetence result in electoral losses for the party or
parties involved? If it does, how large is this effect? In an effort to address these issues,
we create an original, objective index of valence scores for 44 political parties in 46
elections, across nine West European party systems over the period 1976-1998, based on
content analysis of Keesing’s Record of World Events (formerly Keesing’s Contemporary
Archive). The study produces three central findings. First, we conclude that parties that
received low valence scores (relative to rival parties) in a given inter-electoral period
tended to lose votes in the subsequent election. Second, we find that the valence–related
electoral effects we estimate are comparable in magnitude to the effects that previous
researchers have estimated for positional dimensions of policy competition (see Alvarez,
Nagler, and Willette, 2000; Adams and Merrill, forthcoming). In other words, events
such as political scandals, party divisions, and perceived party competence/incompetence
influence election outcomes as much as the parties’ and voters’ positions with respect to
debates about social policy, crime policy, and immigration policy. Third, we find that
governing parties tend to receive more negative valence scores than opposition parties.
This is an interesting finding since previous research examining the vote shares of
governing parties has noted a consistent trend of vote loss regardless of other factors –
holding office has simply appeared to cost votes (Rose and Mackie, 1983; Paldam, 1991;
4
There are some notable exceptions such as Palmer and Whitten, 2000; Anderson, 2000; Chappell Jr. and
Veiga, 2000; Alvarez, Nagler, and Willette, 2000. However, all of these studies focus on the role the state
of the economy plays in affecting election outcomes, or the timing of elections.
4
Convention