Abstract
Have political parties in developed countries conspired to compromise and even pervert the classic model of parliamentary democracy? Have they, in practice, cornered the market in the supply of political representation, and formed cartels that severely restrict the choice available for political consumers – that is, electors? That was the proposition made in the very first issue of Party Politics, in 1995, by Richard S. Katz and Peter Mair. The claim has remained at the centre of debate within party studies ever since – and sometimes beyond it, too (see, for example, Mair, 2013). Even today, nearly a quarter-of-a-century on, it is a challenge to find a book or an article on party systems or party organisation that does not cite Katz and Mair (1995).
The debate continues for at least two reasons. The first is that the cartel argument has evolved, while retaining its core features. In its first incarnation, Katz and Mair argued, above all, that it was the intertwining of parties with the state that symptomised the cartel. Thus, and continuing a tradition in party studies, a new ideal-typical model of party – the state-dependent cartel party – could be added to those previously proposed, such as the mass party, the catch-all party and the electoral-professional party. A decade after the original Party Politics article, however, a ‘political economy’ aspect was appended to the argument, in which it was suggested that economic and ideational trends, in addition to sociological and organisational ones, could explain the rise of party cartels and cartel parties (Blyth and Katz, 2005).
This more comprehensive and ambitious version of the cartel argument is presented anew in Democracy and the Cartelization of Political Parties. In the preface, Katz insists that it is ‘indeed a co-authored work’, because its outline was ready before Mair’s untimely death, six years prior to the book’s completion. Mair drafted three of the seven chapters. The book's style is certainly familiar. There is the usual blend of deduction and induction, of empirical and normative theory; the common-sense use of quantitative data; the rich knowledge of literature and cases; and, above all, the tirelessly comparative approach.
The argument feels even more intuitively plausible now than it did in the mid-1990s. It was clear then that parties, at least in West European democracies, were losing their members. With only a few striking but isolated exceptions, that trend has only continued since. So, too, has parties’ reliance on public finance. The range of serious macroeconomic policy alternatives, meanwhile, is arguably narrower today than it was three decades ago, not least because 19 European countries share a single supranational monetary authority.
Much of Katz and Mair’s book comprises fairly familiar propositions, albeit collected, updated, refined and with a dose of retrospective self-criticism (mainly about terminology). In addition, there is a final chapter, on ‘the cartel party and populist opposition’. The cartel argument was never static or teleological. The original version concluded with the observation that a cartel ‘cannot prevent the emergence of challenges from outside’ (Katz and Mair, 1995: 24) – an assertion that seems prescient, given the subsequent spread of decidedly non-mainstream political forces, often (though perhaps not always correctly) called ‘populist’. And, yes, the cartels have, as Katz and Mair put it, apparently sought to ‘circle the wagons’ (p. 100) against the insurgents. Witness the grand coalitions in, for example, Germany; or the attempt in 2014–2015 by the mainstream Swedish parties – while maintaining a facade of left-right competition between them – to ostracise an eighth party, even though that party had just won a seventh of the national vote. Indeed, Katz and Mair’s allusion to the complicity of the media in party cartels, which reinforces ‘the self-referential nature of elite political discourse’ (p. 174), could have been made specifically with Sweden in mind.
Rather surprisingly, this Swedish example is not mentioned by Katz and Mair. The abortive ‘December agreement’ is, however, described in the very first lines of another current book on the cartel argument – understandably so, since the editors of the anthology, Henrik Enroth and Magnus Hagevi, present Cartelisation, Convergence or Increasing Similarities? as a case study of ‘cartel party theory’, with its tenets assessed ‘empirically, theoretically and normatively’ (p. 16), and with Sweden as the case in question. That brings us to the second main reason for the durability of the cartel argument. It has never been convincingly confirmed or disconfirmed. How exactly do you recognise a cartel party or cartelised party system? The validity of any particular indicator, or combination of indicators, has not won broad acceptance. A really thorough study of a single most-likely country-case of cartelisation, then, has a great deal to be said for it.
Perhaps the most admirable aspect of Enroth and Hagevi’s book, much of which was previously published in a special issue of the Swedish-language journal Statsvetenskaplig tidskrift in 2014, is the clarity of their conclusions. While acknowledging that the evidence is mixed, the editors declare that ‘the (cartel) theory cannot satisfactorily explain the development of Swedish political parties over the period examined in this book’ (p. 206).
That may well be so. But it does not necessarily mean that the cartel argument is thus seriously weakened. For a start, some aspects of Swedish party development, as reported in Enroth and Hagevi’s book, are of questionable relevance. There is nothing much wrong with the interesting chapters on gender balance, internal party culture and parliamentarians’ ‘perceptions of mediatisation’; but it is hard to see their significance for cartelisation. Moreover, even when the data are more salient, it is not obvious that they undermine the cartelisation argument. Arguably, lively intra-party debate and decision-making processes could still occur in cartel parties if the subjects of those decisions had become insulated from the concerns of voters. Similarly, even if parliamentary decision-making has become more conflictual, as the Swedish data suggest, proponents of the cartel argument could retort that these decisions are nevertheless restricted to issues within what another Swedish political scientist has called an ‘opinion corridor’ (Oscarsson, 2013).
Of course, though, you can turn such reasoning around. The evidence marshalled by Enroth and Hagevi might not disprove cartelisation. Yet the evidence presented by Katz and Mair – the decline of party membership, the increasing reliance on public subsidy, the perceived constraints on macroeconomic policy – does not necessarily confirm it. The thrust of Enroth and Hagevi’s conclusions, and that of the introduction to their book, written by Herbert Kitschelt, is that alternative explanations are at least as consistent with the observed data. This is fair enough. On all sides of the debate, what we still lack are observable mechanisms that connect the alleged causes and effects.
In the end, a lot depends on what we want to make of cartelisation. While Hagevi and Enroth refer to ‘cartel party theory’, which in turn ‘consists of two interrelated core hypotheses’ (p. 207), it is notable that Katz and Mair themselves call cartelisation a ‘thesis’, which implies a less demanding empirical foundation. For this reader, at least, their argument is best understood as a work of interpretation, in the sense of ‘an iterative process of observation and conjecture whose goal is to identify an intrinsic coherence to behaviour or events that is not immediately visible to the observer’ (Hall, 2013: 23), rather than one that has the primary objective of presenting testable theory or hypotheses. Perhaps we have to accept that some things in politics – such as, for instance, the signalling through which, according to Katz and Mair (p. 13), party cartels are tacitly constructed – are just hard to get at empirically. That, of course, makes the cartelisation thesis stimulating and frustrating in equal measure – despite the contributions of these two highly stimulating books.