parties” (thereby reducing fragmentation without affected wasted vote shares) than from even smaller, seatless and hence “hopeless parties” (thereby leaving wasted vote shares unaffected) (Tavits and Annus 2006, 77) It also implies that voters of risky parties used different information than voters of hopeless parties. Institutional explanations cannot account for such informational discrepancies since they assume that electoral markets provide uniform information for all parties and all voters learn from that information the same lessons. This discrepancy, therefore, suggests that we look at additional sources of information that influence strategic voting and not just electoral market. Coordination strategies constituted such an additional source information in postwar Germany and thus require close attention. Institutional explanations ignore such strategies presumably because they are still poorly conceptualized and lack systematic, cross-national data.
Second, institutional incentives and efficient markets only reduce fragmentation to the extent that voters choose among well organized, stable and mostly national parties. (Cox 1997, 181-202) The early and continuous vote share of 1st Election Parties in Table demonstrates that such parties developed instantaneously in postwar Germany. The suddenness pose a problem for institutional explanations which – following a functional logic – attribute their formation to the need to form effective electoral and legislative coalitions. (Aldrich 1995; Cox 1987) Such a functional logic would imply a gradual institutionalization of parties rather than an instantaneous one we observe in Table 2. Furthermore, the suddenness can be only partly attributed to the historical legacy of the Weimar Republic. Weimar’s liberal and conservative parties were fragmented, plagued by defections and unsuccessful in transforming themselves from unorganized notable parties into well institutionalized mass parties (Kreuzer 2001, 91-132). Very few of their members stayed in politics after 1945, as only 8.6% of the CDU/CSU deputies in Weimar Reichstag and 13.4% of the FDP deputies. (Schindler 1984, 183-85) With so little personal continuity, center right parties had to reinvent themselves. (Cary 1996) The only interwar parties that resurrected their old organizations were the SPD and KPD. All of this suggests that we look more closely at how starting conditions – that is something occurring after historical legacies but before electoral markets operated - influenced the formation of parties. The effects of starting conditions is overlooked by institutional explanations because, just like coordination strategies, it is not easily quantifiable.