When does turnout matter? the case of Poland

Abstract

THERE ARE TWO CLASSIC REASONS for a concern with electoral turnout. First, a 'top-down' concern centred on legitimacy, authority, governance and ultimately governability. Second, a 'bottom-up' concern about representation, political equality and the 'quality of democracy'. Low levels of electoral participation may erode the legitimacy of a regime, and low levels of electoral participation may lead to an unrepresentative outcome. Both are possible consequences of low turnout, but neither is a necessary consequence. It remains an empirical question whether low turnout does in fact erode legitimacy or produce a distorted outcome. Poland is a particularly interesting case. Brady & Kaplan claim that the Polish double referendum of 1987 'marked the beginning of the end of Communist hegemony in Eastern Europe'.' Ironically, the regime won the referendum by margins that would be beyond the aspirations of politicians operating in most democratic regimes. Solidarity campaigned for a boycott of this referendum-and by today's standards, though not the standards of the 1980s, it failed miserably. Jaruzelski won 64% of the vote on price increases and 69% on political reform-with a turnout rate of over 67% of electors on both questions. But he had set a self-imposed target of gaining the support of 50% of all electors (not voters). So what in any established democracy would have counted as an overwhelming success for the regime was widely interpreted as a 'spectacular defeat for the authorities'2 and contributed to the collapse of the regime. A regime brought down-at least in part-by winning only two-thirds of the vote on a turnout of only two-thirds of electors! Clearly some turnouts matter more than others.