Introduction

SINCE THEIR FIRST OCCURRENCE IN 1979 THE DIRECT ELECTIONS of the European Parliament (EP) have been characterized by low turnout, and 2009 reaffirmed this pattern. The lowest levels of turnout in 2009, 20 percent or less, were registered in Slovakia and Lithuania. In other countries turnout reached higher levels, with rates of over 90 percent in Belgium and Luxembourg (where voting is compulsory), and (for non-compulsory voting countries) highest rates of 79 percent in Malta and 65 percent in Italy. But, irrespective of the actual levels, in all countries – including those with compulsory voting – turnout was lower, and often by large margins than what one would normally find in their national first-order elections – which could be parliamentary or presidential. This is true without exception for all countries, and it has been the case without exception in all EP elections since the first one in 1979. Averaged across all member states, turnout is just under two thirds of what was registered in the most recent national general election before the 2009 EP elections. Obviously, there is some variation in this ratio, and in some countries – the Czech Republic, Lithuania, the Netherlands, Poland, Slovenia and Slovakia – EP turnout dropped to less than half the rate in national elections.

The low levels of turnout in EP elections have given rise to a number of concerns, which have been persistent elements in public debate and scholarly research since 1979. Most of these relate to the causes of low turnout, while the consequences of low turnout have received less attention (but, see references in footnote 3). It is upon the latter that we focus in this chapter. More in particular we examine the consequences of low turnout for electoral representation. It has frequently been hypothesized that some parties benefit from low turnout because their own followers are loyal and active, and therefore turn out in above-average rates, thus yielding larger vote shares than would be the case at higher levels of turnout. By the same logic, parties would suffer from low levels of turnout if their followers would tend to abstain in above-average numbers.