The Electoral College and voter participation rates: Reply

Abstract

In our paper on the impact of the Electoral College on voter participation rates (Cebu la and Murphy, 1980), we argue on theoretical grounds that, in states where there is a dominant political party, the Electoral College system acts to reduce the voter participation rate; these arguments are strongly supported by the empirical evidence. Professor Blair argues that our analysis '(.. . ignores the other side of the coin-the greatly magnified probability of influencing the national outcome accorded a voter in a very competitive state when the electoral vote is close.' In point of fact, however, we do to some extent address this issue in our theoretical discussion (see para. 2, p. 186), although not in our empirical analysis. In particular, we argue that, in states with relatively competitive political parties, the small differences in the political power structure will tend to result in a higher voter participation rate. On the other hand, Professor Blair is justified in his criticism of ou r conclusions. We concluded that the Electoral College should be abolished because it reduces voter participation; however, in reality, such a net effect may not be observed in the aggregate. Specifically, although clearly reducing voter participation rates in many states, the Electoral College may also raise the voter participation rate in other states. In terms of sheer statistics, the final net outcome in the aggregate is theoretically unknown. It is a priori conceivable t hat the aggregate voter participation rate may be lowered, unchanged, or even elevated due to the Electoral College system. As illustrated in our paper, however, the fact that 41 states are dominated by one party or the other implies that, in the aggregate, voter participation is most likely to decline. Our basic conclusion should nevertheless be remolded. It should now emphasize that the Electoral College system unequivocally distorts the geographic pattern of voter participation, artificially raising the rate in some states and artificially lowering it in others. This permits an attack on the Electoral College on grounds that this distortion is not likely to yield net social benefits. Given that the Electoral College induces a voter participation rate

Economics of Voting: Theories and Evidence

“…7. However, in a counter-reply, Cebula and Murphy (1984) argued that severe multi-collinearity between Ostrosky's sole independent variable, and both the income and educational variables reduced the t-value on the independent variable. They dropped the income variable (because of its higher degree of multi-collinearity with the independent variable) but retained the educational variable.…”